Bitcoin Forum

Let's say we see an entity taking over the network. Adrenaline starts to pump into the veins of the larger Bitcoin stakeholders. They go into a primal fight-or-flight response. The non-dedicated stakeholders sell. What do the dedicated do? They buy mining hardware and restore balance back to the network.

If people care about the network, it's safe -- and people do care. There's enough Bitcoin capital floating around to save the network in case of crisis.

Thoughts?

Because when the market is crashing, you should invest into mining hardware!

+1

Care. That is something you do when you have positive cash flow. Don't you realize there is a tragedy of the commons problem here?

Proof-of-stake is the long-run sustainable solution. The current system is a ponzi, but with failure likely only in the distant future so no one cares.

People are doing that right now.

Nope because people want to continue to own Bitcoin.

http://www.youtube.com/watch?v=MLirNeu-A8I

If there is a 51% attack (and I don't think there will be) there will be no warning. The attack chain will be built in private and contain thousands of double spends across hundreds of individuals.  It will be published and instantly replace main chain creating economic chaos, plummeting bitcoin values.  If done right the attack will come right after difficulty adjustment so say BTC declines 90% in value.  It is now massively unprofitable to mine so miners quit this pushes the 2 week adjustment window out to 4 weeks, 6 weeks, 8 weeks or more.  Plenty of time to double spend the network into oblivion.

The idea that people will not only run their rigs as massive loss but also buy and deploy new rigs is naive.  It is like saying recessions won't happen because everyone will do the right thing, dip into savings and increase aggregate spending 3%-5% preventing the recession and resulting layoffs.

The latter scenario will never happen because recessions are caused by the cores of the system in the first place: Controlled money supplies.

It's not naive to think the people will act when the people have direct control of the system.

No, The Federal Reserve is not the will of the people.

Tragedy of the commons.  In a recession the best thing an individual can do to protect themselves is to cut back spending.  That applies to both individuals and companies.  The tragedy is that aggregate effect is the very reduction in aggregate demand they were fearing.

No individuals won't throw money into the fire hoping to stop a 51% attack which has already occurred and by definition indicates the attacker is extremely well financed and committed.   "Defenders" could spend tens of millions of dollars AFTER the fact with Bitcoin crippled and the attackers could spend EVEN MORE and defenders lose not only what was already at risk, and not just ongoing electrical costs but all the new capital put into play at the riskiest possible moment.

The idea that individual miners won't act like individuals and pull the plug to limit their own losses is well frankly naive.  We aren't ants.  We are people and people look out for #1 first.

Not everyone's selfish desires are the same. Some people's #1 is Bitcoin as a whole.

It's mine. I assume it is other's as well.

In addition, there can be long-term profit from sustaining the blockchain even in dire scenarios. The incentive on an attacker's end can usually only be so much. The defender's will usually have more.

You're right, people are selfish but not everyone has the same target desire.

Sure some may sink money into a bottomless pit but some won't.  51% attack by definition means the attacker has >50% of network capacity.  Any "good" hashing power that quits only increases the attackers superiority over the remaining network. If say 2TH/s of good miners quit then the remaining defenders would need to deploy 2TH/s (@ $0.75 per MH = $1.5 mil) just to get back to where they started at <49%.  Then they would need to spend an unknown amount more to overcome the attackers margin of superiority.  All while working for essentially free and knowing there is the chance any capital deployed will simply be lost.

Lets face it a 51% attack is an 8 figure operation.  If someone is willing to spend $10M+ they likely will be willing to spend a little more and make sure they end up on top.    If done right the true strength of the attacker won't be readily known.   Defenders won't know if the attacker has 51% of the network of 99% of the network.  How longer the attacker has been planning it won't be known, and how many blocks the attacker has mined in advance won't be known either.  Essentially you are proposing that facing a high risk of catastrophic loss and no way to define the chance of success that people will keep dumping money at the problem until they "win".  That is unlikely and even if it is likely it may be futile.

Say the attacker started 4 months ago and has mined an attack chain not only to the current block (double spending over the course of the prior 4 months) but an additional 30 days worth of blocks into the future.  The defenders are behind so far that it may take months if not years to pull ahead.  That could be happening right now with no way to verify.  So when the attack comes you have no idea if the attacker is barely ahead of TH/s ahead.  How much hashing power do you need to add to pull ahead.  Even if you matched the attacker in hashing power they have the time advantage of working months with the larger hashing power so you would need to build out large enough to overcome that.  Worse the "even that" is completely unknown.  Maybe they built a 2GH per chip ASIC and have 80TH/s already deployed not just today but 4 months ago.   The defenders increasing hashing power of good network from 9TH/s to 11TH/s would be as futile as a bunch of cavemen doubling the number of spears when facing a nuclear strike.

If we're playing the "if" game, then if that attacker started 4 months ago, then their competition would have a 2GH per chip ASIC and have 120TH/s already deployed not just today but 3 months ago and so on. I actually do think this is a plausible scenario.

I would advise that you leave thinking to people with the proper equipment. Your job is to listen quietly, nod, and ask polite questions.

For example, if you don't know what tragedy of the commons is, then you should ask "What does that mean?"

"The Tragedy of the Commons" only exists with flawed systems that can't meet the desires of the people; the common good, if you will. If that common good is never met despite the property owners involved still having proper incentive, maybe the "common good" isn't all that desirable?

In this case, it is.

I agree with Boss. The other two guys are complete assholes.

Bitcoin is run by volunteers, like wikipedia. Not everybody is driven solely by money.

Besides, later on the miners will be paid by fees from transactions. But even without fees, I would contribute to the network. I think this is something that is in general very difficult for many Americans to understand.

If the market drops, mining difficulty decreases (correlation verified) and I resume mining.
My stake in bitcoin is not money.
Bitcoin is a movement for freedom.

Isnt there restore points saved? So a restore of the network to before the attack is possible?

Making such an attack somewhat meaningless?
You will temporarily crash the market, but when its restored people will get even stronger confidence?

Yes this is true. I began mining as the price went from 32 to 2.

Actually I began when the price was in the 3's. Some people will always move counter intuitively and this is hard to predict.  And I could make an economic argument that it makes sense to start mining when the price is low. Those mined coins could appreciate more so over time and they would have been mined when the difficulty was low.  It just depends on your viewpoint.

I too would mine at a loss to protect the market. Its about a principle.  and my investment.

yes, good point.  there are checkpoints put up along the block chain as it gets longer where the blockchain history becomes hardcoded.  the last one was at 140,700.

that means that even with 51% of the hashing power of the network, the attacker would NOT be able to overwrite any tx's prior to that point.

also, Gavin said he will be moving that checkpoint forward very soon.