Hacker News Reader

Comment by jobeirne | Hacker News Reader

Folks are missing an important point here about the incentives of a pool majority (that I recall Satoshi mentions in the original whitepaper[0]): even if a particular agent has a majority in the mining network, it isn't in their interest to upset trust in a system that they have such a large investment in.

Where would the sense be in taking the risk to undermine public trust in BTC for a short-term gain like modifying a few transactions when you have such a large stake in the system?

[0]: http://bitcoin.org/bitcoin.pdf

View on HN

Here is a thought experiment :

Say the pool operator is also the owner of Bank A (which has nothing to do with BTC). Now, a competitor of Bank A, Bank B, just announced that they will support BTC. The pool operator can now decide to "destroy" bitcoin and therefore cause severe financial loss for Bank B, leading to bankruptcy. If putting Bank B out of business results in gains (in USD), for the pool operator, that are greater than what he just lost in bitcoins, then it might be worth it.

Of course, this is very hypothetical, but it's not hard to come up with similar situations where Bank A is "State A" and Bank B is "State B".

Thanks for the concrete counterexample. I still find it dubious that the owner of Bank A would be willing to destroy a stream of income as considerable as a 51% stake in BTC just to preempt a competitor in another market. That seems too self-sacrificial to be practical.

Also, you're missing the point that Bank B's successful adoption of BTC would also benefit Bank A's owner, since that adoption would be a boon for BTC, which Bank A owner has a very large stake in.

It's all a matter of the right price. Bank A doesn't own 51% of bitcoin, it has control over 51% of the mining pool. As a rational agent in the bitcoin-only world that's worth SOMETHING but not billions. As a rational agent in the total world it might be worth sacrificing $100mm on the pool if it can cost your competitor $10b.

Good points. What I wanted to point out is that the typical arguments about an agent's incentives to not destroy BTC usually fail to take into account gains outside of the BTC world.

This is like asking why a counterfeiter would undermine trust in the U.S. $100 dollar bill in order to get a big short-term gain.

I know there is a big spirit of brotherhood about doing what's best for BTC, but if your prototcol is to survive the real world, you have to anticipate people more interested in their short-term gains than global harmony. Particularly if the protocol will involve money. Particularly if the protocol is trying to be money.

A counterfeiter sees far less benefit from the continuation of the USD than does someone who owns 51% of the BTC mining infrastructure.

https://bitcointalk.org/index.php?topic=327767.0

I dunno, maybe to double-spend their coins? GHash.IO has been accused of double-spending in the past. Reaching 51% will enable them to double-spend their coins.

If GHash.IO really cared about BTC, they would stop before they reached majority stake in the Hashrate. All other pools in the past have purposefully gimped their hashrate in these situations.

A pool reaching 51% is undermining the trust of the BTC system. Period. GHash.IO apparently doesn't care however.

---------------

Furthermore, as the thread I linked to suggests, GHash.IO attempted to double-spend against BetCoin back when they only had 24%. This isn't a theoretical problem, these guys actually are trying to double-spend bitcoins.

I think they've all but admitted that they were behind the double spending actually, they just blamed it on a rogue insider: https://bitcointalk.org/index.php?topic=318010.msg3590355#ms...

>All other pools in the past have purposefully gimped their hashrate in these situations.

This has even occurred with altcoins (which are often widely believed to be pump-and-dump schemes) where a pool gets too close to 51% and voluntarily makes an effort to reduce their hashrate (either by blocking new signups, issuing fewer work units, etc.)

What if somebody incentives the pool operator with a rubber hose?

Reply on news.ycombinator.com