Bitcoin Forum

No institutional investor is going to put millions of dollars into something that can be killed off by a few douchebags like Ghash.io. Think about it...

+1

The 51% question *is* answered for those who care to understand its true ramifications

YOU CAN'T FIX STUPID! It's a disease.

I will admit, I sold some BTC when this came about. I figure the only way to solve it is go there (>51%), come back and learn from the experience. Even if there is no attack, private mining giants like ghash.io remind me that even with such a brilliant system as BTC, there are still humans involved. Truly free market will result in some entity writing/updating the blockchain, it worries me as an idealist...

can you link me or summarize what you're implying?

https://en.bitcoin.it/wiki/Weaknesses#Attacker_has_a_lot_of_computing_power

The 51% attack question is already answered.  Big corporations have no risk from accepting BTC since a place like Amazon would have an employee tasked with the job of monitoring hash rate and halting the acceptance of BTC payments should the hash be outside of what they're comfortable with.  The bigger a business is, the less risk they have due to practices like this.  I'm sure there would be 3rd party services that small business could pay small fees to as well to constantly monitor potential problems should they get involved in the BTC acceptance business.

FYI, pool Ghash.io has mined 48% of the blocks in the last 24 hours. On Sunday, it was 45%. It starts to be hazardous.

Than create a pool like ghash. Everyone complains about it but no one comes with a pool as good as ghash

What is so better with Ghash?

the answer to everything.

There are several pools that are better than them, just not as flashy

or as lucrative. I'll switch to a 0% fee pool that finds multiple block in a day in a heartbeat.
Pick a pool and lets move.

There's nothing hazardous about that. Bunch of hysteria over a practical non-issue.

Basically, someone with 51% can reverse a few transactions and prevent some people from spending money.  If you see greater than 51%, its best not to send or recieve any bitcoin at the time.  If your coins are buried more than six blocks deep, you should be fine.

Frankly, I would have to see one single entity control more than 51% of the hashrate for a lenghty period of time AND people complaining about their transactions being reversed or not getting included in a block for me to panic.  I realize the possibility for a major problem is always there, but of course you could have just as bad or worse problems at any given time with fiat.

Would Satoshi not have envisaged this problem?

He did and outlined it as a MAJOR RISK

This is why he is constantly crapping on about securing a DISTRIBUTED network

+M'fking 1000.

Idiot n00bs refuse to believe that this question has been discussed (and summarily answered) over and over... for 4 fkn straight years now.

Trolls will be trollin'.

From the wiki:

"Since this attack doesn't permit all that much power over the network, it is expected that no one will attempt it. A profit-seeking person will always gain more by just following the rules, and even someone trying to destroy the system will probably find other attacks more attractive."

I've read from this link: http://learncryptography.com/51-attack/


I guess 51% attack could drive away the investors  :-\
Like maybe two companies doing a multi-million dollar deal using bitcoin as payment gateway but their transaction cannot proceed because the network is under 51% attack. They'll be forced to avoid bitcoin.

So, I was wondering, does bitcoin have the capability to ban those miners/people doing the attack? Let's say some entity could maintain a 51% hashrate and they continued attacking the network,  what could bitcoin do? Will bitcoin just be a sitting duck waiting for someone with greater computational power to appear and save the network?

lmgtfy http://gavintech.blogspot.co.uk/2012/05/neutralizing-51-attack.html

That assumes they are in control of their pool and not some third party.

The tired argument that the disincentive to the 51% attack is so great that no one would consider it... is a flawed argument that comes with the assumption that there are only rational actors on stage...

We all know that there are men who just want to watch the world burn... that 51% attack is coming, and it needs to be addressed.

 Thank you and Thank you, I was going to post the exact same thing.
Just because the pool operators have no plans of a 51% attack doesnt mean the pool cant be hacked and someone else does the attack. We could bounce back after such an event but looking at other coins who have been 51%'d im not sure....

You're an idiot OP

It is a problem. It doesn't matter if there is an easy technical solution or if we can recover from it, what really matters here is public perception. It must not happen in the first place.

Remember when bitcoin hard forked in March 2013? Remember how incredibly quickly the community came together to remedy the issue? Remember what happened to BTC/USD price right after?

Do you think miners wouldn't leave ghash.io quickly and permanently if they executed a 51% attack by their own virtue or that of a hacker? Do you think in the very rare event that a pool did execute what would be limited to a brief 51% attack there would be any dire, permanent consequences to the integrity of the protocol? Or do you think it's more likely that the bitcoin community would once again demonstrate its resiliency?

"We're not going anywhere..." until the 51% question stops being asked. Seriously, it can't be that hard to understand.

It will stop being asked when it stops being a problem. Seriously, it can't be that hard to understand.

But it's not a problem, seriously. Stop.

Are you telling me that the fate of bitcoin resting resting on a single entity who can fork it on a whim, is not a problem?

Instead of wasting your time begging me to stop, you should beg GHash.io to stop instead.

+1.

What's the difference between bitcoin and centralized payment systems (visa, mastercard, paypal) then?
The de-facto single payment processor is not what we all  are expecting from bitcoin. Even if our money is safe (for now).

Meanwhile, GHash.io is at 47%. I bet it will reach 50% in a few days.

one of the main points is Bitcoin is decentralized. Allowing one entity with a large percent of the hash rate defeats this and goes against the whole idea. I really have a hard time understanding why people dont see this as bad.

Because if you looked into the "issue" it's not very impactful (wouldn't do major damage to the system) and completely stops taking effect after miners switch pools immediately following the attack.

The fact that they hit 48% tells me the average miner doesn't much care about it. It should never have gotten that high.

Because it's a non-issue. And if it becomes an issue, they will obviously move, as their profits rely on the success of bitcoin.

The system will become the same as visa/mastercard/paypal. Isn't that a major damage?

No, it wouldn't. Stop trolling. As previously mentioned the effects wouldn't even last long.

I don't see why the centralized payment system "Bitcoin" (owned by GHash inc.) wouldn't be the same as Visa.

Because that's not what an individual with 51% mining power could do. Educate yourself.

Will Ghash.IO be able, for example, to block all transactions from Russian IPs because of US sanctions? Or because the owner of GHash is from Ukraine.
Simply not include those txs into blocks.

No because within 10 minutes Ghash.IO would lose at least 5% hashing power due to panic ending the 'attack'

No, because it will not become the same as visa etc. IF this becomes a problem, it will resolve itself quickly. If one central authority owned enough HARDWARE to execute a 51% attack then it would be slightly more disturbing, as they could have solely malicious intent. Their is no profitable model for catastrophe in the current situation. In fact, there is no model for catastrophe in the current situation. Miners will move if a REAL problem arises, but it almost definitely won't.

Educate yourselves via one of the most knowledgable and prominent voices in bitcoin, Andreas Antonopolous:

http://www.youtube.com/watch?v=bTPQKyAq-DM#t=2942

The problem is that miners won't care about US sanctions against Russia, or about Ukrainian people hating Russians.
Miners even won't notice that some transactions are not being included into blocks.

There are a lot of threats in this kind of situation, whatever Andreas says.
And no one does anything real to prevent this centralization in the future.

The system should not rely on people's responsibility. The system should be designed to work even if everyone is too stupid to change anything on his computer.

Pretty sure you're trolling us now. Have fun with that.

Just a quick note: This attack is possible with anything greater than 50%.  51% is not required.

GHash generated 7 of 12 blocks since 0:00 UTC today. It means 58%.

THANK YOU. So tired of the fud about this.

For those who believe news controls the markets, that's good enough news as any to attribute as the reason behind the recent price drop.

Hadn't heard about coinjoin. Auto-mixing sounds like a great idea.

just cut the cables .. is it a deja vu scene ?

https://www.youtube.com/watch?v=lcwCnxOrA0M

I guess they better start double spending then!! I mean, this would be the time. Right? Right.

Me too. Glad you took the time to watch that section. The whole video is great from start to finish if you have a couple hours to set aside.

The attack is about having >50% of the hashrate, not necessarily the generated blocks.  Block generation is completely random.  It's very possible for 1 person to generate 100 blocks in a minute.  Very improbable, but completely possible.  It's the hashrate that's important.

How does the pools' hashrate get monitored? Is it based on the data that the pool provide? What if let's say pool1 have 40% hashing power then pool2 have 11% then they combined their hashing power for a brief period of time? Will it get monitored?

To be honest, there is no way to really monitor this.  If a few pools get together and decide to try to double spend, it would happen.  It would have to be a self destructive act, so it would likely be a hacker that was controlling all of them simultaneously.  It wouldn't be a rational hacker, as it would be more profitable to steal the BTC from the pools than force a double spend attack.  It would have to be a hacker with the psychological makeup of the Joker.  Once this attack happened, the Bitcoin Foundation would quickly move to reverse it and implement countermeasures.

Well I think this opinion might be partially correct that said I read a few threads on Gigahash abusing its power when it last had greater than 50% of the network by reversing bets.
That said the network should correct itself if it is abused again so a non-issue but with a caution involved in it.

That said there is more incentive to leave the system working as it is than to double spend on it and lose more than you would gain

So, is this already implemented on bitcoin's code?

why not track down miners and destroy their equipment in a criminal act?

or at least the miners who are those imbeciles mining as a horde under the same mining pool. or maybe it would be possible to disable centralized mining from bitcoin altogether?

If bitcoin 0.9.2 was to come out with the hard fork so that newer versions no longer allow centralized mining I would rather support it and have everyone forced to use P2P mining.

I think the best solution would be if we as a community decided that each mining guild should not exceed 10% of the total network rate. This way people are still ensured regular blocks and it would be much harder to get enough groups together for a 51% attack.

I am not quite sure how you would get pool owners to accept a change that would negatively affect them.  At best they would just subdivide their pools into smaller pools and carry on, possibly obscuring their ownership.

Not the pool owners, the miners. And the second problem would be a matter of not signing up for someone with unknown owners. But of course that won't happen until we have an actual 51% clusterfuck.

no because anyone smart enough to warrant trying it will already understand why its not worth it. but *just in case* there are stupid people in the world with more money than sense (i know). that thread gives some reassurance that its not the issue people think it is.

whats more telling is the way people seem to think this is a black an white issue.

Setting aside transaction manipulation, which is largely a non-issue given the nature and transparancy of the bitcoin protocol.

In terms of block reward hash rate doesn't guarantee anything. hash rate as the fuel to the fire that is block generation. block generation is what gives you the power to manipulate the chain.

Having 51% doesn't magically give you guaranteed control, just as 49% doesn't mean you can't possibly manipulate it.

The hash rate is a reflection of the probability you can solve the next block before anyone else. 51% is marginal, its basically still a coin flip, just like 49% is.

With 51% The chance you can solve two consecutive blocks is 0.51^2 = 26%
The chance you can get 6 blocks is 0.51^6 = 1.7%

Anyone with the most base understanding of how things work can see that with 51% of the hashrate you are going to average 75BTC per 6 blocks.

Over time you are averaging 12.5 BTC per block.

Or you can get 150BTC 1.7% of the time by trying to build your own 6 block longer chain before anyone else.

Over time you would average 0.43BTC per block.

To make it worthwhile, you would need to get your 6 block chain in 50% of the time or more. To guarantee that you need 89% of the hashing power (6th root of 0.5)

If you have 89% of hashing power though, you'd still be better of just hashing. because you would average 22BTC per block.

In a way its good that idiots with a lot of hashing power might try this, because in the long term it means their hashing power wouldn't count, and any remaing honest miners get a bigger piece of the pie ;)

Yeah but remember, people are dumb. Most don't do numbers. Look around this forum, how many people give the impression that they understand what they have thrown money at? I could name a few, but most don't fall in that category. If there ever is a 51% attack that gets even a little bit of publicity there will be panic.

I can answer it with a question: Why care? By the way, I know who John Galt is.

No, I can answer it better. The problem with the 51% attack is the name. It's too catchy.

all the ones who are the first to chime in here saying who cares it isn't a problem will the first ones crying they lost their lifesavings.

As if this exact same FUD wasn't circulated only 6 months ago in December/ January.  ::)

Last time the network hard forked, miners acted swiftly to resolve the issue. Shortly after the huge (but quick) panic, we saw the beginning of a massive rally. The ones who panicked lost. The ones who understood the issue made a lot of money.

In the incredibly rare and stupid event Ghash forks the network, you can bet we would see similar action. The weak would panic, the miners would move, the network would be fine, and the strong would pick up some cheap coins.

be careful to avoid actually discussing any facts, and make sure to only use hyperbole, sensational language and speculation on this forum. its's what keeps it ticking!

I don't think any investors care about the 51% issue because everyone knows that it's not an issue. It's never going to happen. The person it would harm most is the person who controls 51%.

why is it always about ghash - go beat up on BTC guild to cut the 3% fee to something vaguely reasonable. Then maybe miners will stop pointing at ghash

LOL How true.  ;D

I don't really care. Let it crash. More ppl mine on ghash the more blocks solved = faster money for me.

The question has been answered : the risk is utterly low

No, it doesn't.

and what happens when a pool hits 60% or 70%? no risk still?
never happen?

The March 2013 fork was much scarier and the core devs and miners had the prob fixed in an hour.

If you really want something to worry about, worry about ECDSA being broken.  If someone could crack public addresses within a reasonable amount of time AND without forewarning, then things might get interesting.  Of course this scenario is next to impossible also. Anyone able to crack ECDSA would have to make a decision of whether to steal some bitcoins or set off some nuclear warheads first.

If someone is scared of this then don't join in. We really don't need institutional investors anyway. Especially those who don't understand what a microscopic risk this is. 

I think bears are still hoping the market would drop on this 51% scare stuff (....this again???) -- wonder if they're right. No big moves yet....

This has already been priced in.

A few newbs panic selling.

im not sure there is any point in you using the question mark when you write posts, because the question mark implies you are asking a question.

This in turn usually means you seek an answer.

And yet, having been provided answers to these and other questions you are still compelled to post, asking the same questions.

Did you read the answers to your previous question?

Did you understand that the more hashing power you get the closer you get to being able to control the block chain, but that the potential benefit of doing so is far outweighed by the number of coins you get if you just hashed?

Can this be put to you any more simply?

As mining rewards are halved, this argument becomes weaker and weaker.

Ah, no. It's still the same percentage of the total mining rewards we are looking at. And transaction fees will continue to go up over time, and those are most definitely related to how much people trust the system.

Right.  But a potential attacker isn't interested in percentages.  He is interested in raw dollar amounts.  If he can find a way to get more money with an attack than with mining, then there could be trouble.  Add to that the potential for an irrational actor, someone with the psychological makeup of the Joker, and this isn't really something that should be ignored, imo.  The Foundation should proactively work on a solution.

I think the point you seem to be missing is: *he can't*

Looking beyond the irrational actor argument, for which you haven't provided a counter-argument, the NSA could decide that Bitcoin is a threat.  They could have a team of hackers break into a mining pool that controls >50% of the hash rate and perform the following operation:

1) Transfer all of the pool's coins out to an exchange, sell them in small chunks for USD, and wire to an undisclosed location.
2) Initiate a double-spend attack on the coins that were sold, thus stealing them from the wallets of the general public, and causing Bitcoin to fail in the eyes of the public.

The Bitcoin Foundation thus has 2 options on how to respond.  They can either:
1) Leave the new blockchain in place.  Have an angry population that was stolen from.  Bitcoin loses confidence and dies.
or
2) Revert the blockchain such that the double-spend never took place.  Consumers keep the Bitcoin.  The largest Bitcoin mining pool goes bankrupt.  Miners do not receive their rewards and become disillusioned with Bitcoin.  Hash rate suddenly drops substantially.  Bitcoin becomes open to further >50% attacks due to low hash rate.  NSA repeats the operation until Bitcoin is dead.

Or... the Bitcoin Foundation could be proactive and fix this before we have an attack in the first place.

Just dont mine at ghash.io and make others not.
Easy!

This doesn't have to be the NSA either.  Any sufficiently advanced government could do this.  China would have a good motive.

It's really not something that can just be ignored. The risk right now might be low, but there is nothing theoretically stopping a government or the world banks from setting up enough hardware to get, say, 99.9% or more of the total network power. It's in the billions for that much, and that's even assuming they do it by honest means. Given what's at stake, it's peanuts. If there is a solution available it should be implemented.

Well if that's what they wanted they could just out right ban Bitcoin, but that's not happening.

It would cost 139943424$ just to buy the hardware to control half the network.


So yes, it is something that can be ignored because it is not a problem or an impending issue.

If you wanna worry, worry about bans and regulation.

51% is that not that big of a threat because it is something that can be recovered from with a hard fork and in the future the vulnerability will be fixed alltogether with some patch.  The real threat is EC being broken.

There is no time limit. And there is no way to know that someone is not gathering hardware right this moment. And again, peanuts.

Why wouldn't we fix this?

I think you, like many, overestimate the ability of the NSA and underestimate the resiliency of the bitcoin network as far as a 51% attack goes.  Miners would leave compromised pools as soon as they detected a problem, probably before the 51% threshold had even been reached.  The attack you describe would be a long shot at best with minimal success, if any.

Like I said before, the NSA would have better luck and inflict much more pain if successful by trying to crack ECDSA, and they probably are trying to crack ECDSA.  The fact that they haven't cracked it yet speaks to its resiliency.

We wouldn't know if they did. But it should be possible to switch to another protocol if that happened.

I can assure you, there are no rich entities looking to spend over 100 million dollars to destroy our monopoly money. The thought is asinine and I can unequivocally say that the biggest danger to Bitcoin will not arise from any one of these crazy theories. Fear regulation.

Also we can just hard fork.

Maybe they don't want to fix this YET.... because they're reserving this for the future price dip  ::)
 ;D ;D ;D

In response to the people above. No it will not take 100 million to buy up the hashrate. It might not take 20 million to buy the majority of hardware to mine. This is because the people who make the asics chips (Taiwan etc) can either themselves or sell to others to hash. They can sell at a fraction of what knc or something sells to the public.

What I currently find amazing is that the devs feel that "dust" was more important than a pool having X amount of hashing power.
Where a pool can deny dust or accept it but a pool won't deny extra hashing power.
I don't get it.

Cause manufacturers would sell at a loss, right?

Those ASIC miners are being sold like a hearing-aid.. Doesn't cost much to manufacture but sells at a high price

If they desired, the Chinese government could order the manufacturers to produce any quantity they wanted.  In China, everyone just does as the authorities want.  So if the day comes that China actually wants to harm Bitcoin, they will have all the hashrate they want.  Sure, it can be fixed after the fact.  But why not be prepared?  Why not fix this before the fit hits the shan?  Why not shore up investor confidence?  Speculation is just as much (if not more) about PR and image as it is about fundamentals.

Well, I was reading somewhere about one of the older pools deep something that had 80% of the hash rate at one time (I am vaguely quoting something I read on here so don't hold me to it)

What they basically said was that if people saw any double spending, people would move to different pools. Look at the charts now, I am sure people would switch to lesser pools to unite against majority pools or something of this nature.

Again, we have not been through a malicious attack (double spends that were confirmed etc) so there is no telling what would happen.
---------------------
https://bitcointalk.org/index.php?topic=407217.0
edit, i read it here
last comment on first page by rocks and I quote:

"I admit I don't like this given GhashIO's history.

But when I first got into bitcoin, Deepbit (remember them) was 80% of the hash rate and they had 80%for a long time.

Deepbit at 80% didn't destroy the network because: 1) that is not in a pool's interest and 2) any attempts are highly public and would result in the majority of your subscriber base immediately leaving.

A pool is only as strong as the trust individuals place in it."

I agree with his statement

Or China could officially 'ban' Bitcoin.

But sure yeah let's believe the government will spend multi-millions to destroy our monopoly fun.

There was never a time when a single pool had 80% of the hashrate.   At one time deepbit did exceed 50% briefly and there was similar gnashing of teeth and wailing about the end times for Bitcoin.  Of course some miners moved and deepbit lost popularity to other pools and life went on.  The reality is being the #1 pool is a license to print money.  The higher their hashrate the more their fees.   Betray the public trust and it is gone forever.  There is nothing you could possible double spend that would make losing that a good idea.

Thanks for clearing that up. I did quote that from someone else (and I was not around when deepbit had 50+%). It would make sense that the same people would be nervous about that (similar to ghash now). 

Also I am fairly certain GHASH.IO doesn't have 50% of the hashrate.  The 1 day chart can be very misleading.

Lets pretend a pool, pool x has 40.3% of the hashrate.  This means that in 24 hours it should solve 58 of the 144 blocks.  The network has no way of knowing how much hashrate pool X has, only how many blocks it solves.  The problem is that the variance even over 24 hours can vary significantly.  Lets look at the probabilities the pool will find more than 58 blocks per day when looking at a 1 day and 4 day period.


The first thing you will notice is there is only a 6% probability, that the pool will produce the exactly the 58 blocks expected in one day.  That is the only time the computed hashrate will reflect its actual hashrate.  When a pool does better than expected miners call that luck so you using that terminology the pool will be lucky, solve 59+ blocks and thus appear to have more hashrate, about 47% of the time.  It will also be unlucky, solve 57 or less blocks, and thus appear to have less hashrate 47% of the time. 

Even variances of 25% or more aren't that rare in a one day period.  This hypothetical pool, which remember never has anything other than exactly 40.3% of the global hashrate, will have a 3% probability of appearing to have 51%+ of the network hashrate.  This means that by only looking at the last 24 hours, the pool will appear to have 51% of the hashrate about once a month.  Looking at a 4 day window reduces the probability that you are just getting excited about variance.  It will on average only show a variance of 10% once every two months and will only show a variance of more than 20% once every five years and will only show a variance of more than 25% about once a century.  An even longer period is better over 2016 blocks there is only a 3% chance the pool will deviate by more than 5% (will appear as 38% to 42% of the hashrate 97% of the time).

Why did I pick 40.3% for this "pool x"?
https://blockchain.info/pools?timespan=4days
http://blockorigin.pfoe.be/chart.php

It doesn't matter if deepbit topped out at 80% or 55% or 45%.

The main point is that that a large-ish pool has not been an issue in the past and won't be today either because: 1) it is not in a pool's interest and 2) any attempts are highly public and would result in the majority of your subscriber base immediately leaving.

If today's largest pool made something as simple as a 2 block re-org that invalidated any transactions, their hashrate would drop to near zero within days.

Again, a pool is only as strong as the trust individuals place in it. With a public ledger this trust is not blind but checked and verified constantly...

Hello rocks, I quoted you because I believed what you were saying. I also agree with what you wrote above as well. I do think people would simple hash at another pool. Seems pretty easy to do. Even some manipulation would cause the big miners to protect their investment.

And also, I do not think ghash would try anything malicious. They have too much going for them with their cex fees  ;D

=======

Exactly, and same as this statement, it is just as true that the NSA/government is only as powerful and relevant as the people allow it to be. People have been so conditioned by this evil empire as a single organism like "god" - with the same sort of power, all knowing, all seeing, all powerful, so any time they hear about the government doing this or that, they freak the fuck out like somebody got god all up in a tizzy and now the fire and brimstone fireworks are coming and we're all going to be punished for disobeying the great and powerful Oz.

In real life, there is no god government, no "illuminati" crap...there are only individuals making up agencies in a larger network of other agencies.  It's not a single entity that controls the world. It's a lot of clusterfucked smaller agencies with the advantage of having a system that is so fucking ridiculously filled with red tape and loop holes that no single individual or even chain of command even knows the whole system inside out. It's all just PEOPLE at the end of the day. It looks spooky only because there are so many channels to deal with it's overwhelming...and it looks like genius to the outsider being victimized by it. They got this sewn up to where we *can't* comply even when we truly want to...they've made us all slaves. We surrender.

No. In real life they don't have any centralized control either, the one arm has no fucking clue what any other arm is doing, and one arm's rules conflict with some other arm and nobody can get shit accomplished because there's no end to the run around. This is NOT by design of some brilliant 300 year old secret cartel else they'd have been murdered in their sleep already.

This is a result of PEOPLE not being capable of large scale concentrated coordinated efforts. Plain and simple.

Small groups yes, but THE GOVERNMENT or THE NSA is not an entity. It's just a large number of PEOPLE with conflicting ideals, goals, rules and red tape that make it such a clusterfuck of horse shit, nobody has time to deal with the overwhelming string of endless nonsense from one branch to the next.

The NSA isn't the spy god in the shadows. It's just PEOPLE getting other PEOPLE to give them information. If the other people decided not to, the NSA would grind to a halt and look like as much of a joke as the pope looks these days. They're only as powerful as other people ALLOW. Most of the success of the NSA has come directly from the reality other people are lazy and unwittingly making accessing them easy. When they wise up and start implementing various blocking security, the NSA loses enormous power...same for any other branch of the government.

Bottom line with all this fear mongering on here...if the NSA *could* do it, if any wealthy billionaire *could* do it, why HAVEN'T THEY done it already? Bitcoin's been around going on 6 years or so right? The more time passes, the less bitcoin are mined, so it'd have been far more worth their effort to pull this early on...but they didn't.

A lot of the fear mongers on this issue don't seem to properly understand MEMG. I'm not even remotely pretending I fully understand the 51% attack in entirety but I do understand it enough through reading more on it to know that MEMG applies in this case making it unlikely.

MEMG = Minimum Effort, Maximum Gain

Anyone with significant resources to successfully implement a 51% attack would not even remotely near a break even point in gains, and would then be blocked and obsolete as the responders took control. Nobody who is capable of making the money required to obtain these sorts of resources is stupid and irresponsible enough to pose that sort of risk to their OWN financial means. Their goal is making money, not losing it. For them to spend thousands of dollars in resources for a momentary minimal gain is insanely stupid...maximum effort for minimum, short lived temporary low gain that doesn't even provide a recovering of their investment.

That's WHY nobody would actually DO it...just because they CAN do it doesn't mean it's going to happen anymore than any one of you CAN stand in front of a speeding bullet train and potentially survive and sue the train company for damages and make a million dollars you'll be forking over to a hospital the rest of your life...but only a dumbass would actually DO it.

And if one dumbass does do it, it's not bringing down the train companies, the medical institutions, or the legal profession. It's just weeding out the next winner of the Darwin Awards.

Well said. For the government types to know all the publics holdings and transactions, the public has to give that information. I don't think the public will want to do that. They will bring forward the sacred "no".

*golf clap*

It's time for more decentralized mining pools like P2Pool to be written and adopted. Perhaps somebody will write a new protocol which fixes whatever is unattractive in p2pool and adds the benefits of stratum.

Time for the 51% question to be answered?
https://blockchain.info/pools?timespan=24hrs
GHash.IO over 50%.

It does look scary. I do not understand why GHash.IO owners will not take some action to curb this? It is also against their interest, they are miners, why do they contribute to eroding trust in btc?

This thread will resurface like clockwork.

However, OP has a valid point: this is an issue that needs to be addressed before bitcoin transitions from a novelty-payment scheme, to a globally utilized financial tool.

Ghash is very good at marketing and has likely attracted a lot of "hands off" type investors that want to look at the pretty numbers. Ghash also offers merged mining of other SHA coins, increasing the EV of mining with them a little bit.

Even if ghash were to control 51%+ of the network does not necessarily mean they will carry out an attack.

+1
i dont think they would. makes no sense to mess up all for a few double spends b4 we catch on

Looks like the community is answering this 51% attack rather quickly and violently. 24-hour block discovery is down to 36% for GHash.
Things that have happened since the attack:

+ GHash was DDOS for a moment last night. (http://www.reddit.com/r/Bitcoin/comments/2854n4/cex_ghash_was_ddosed_last_night/)
+ BitFury Pulls 1PH/s of Mining Power from Ghash.io (http://www.coindesk.com/bitfury-pulls-power-ghash-community-uproar/)
+ P2P (https://i.imgur.com/gpFXnkH.png) reaches 1% network hashrate. Discovers blocks (see 4-day timechart (https://blockchain.info/pools?timespan=4days))

In spite of the temporary relief, I suspect the GHash reduction is superficial, and will guarantee to surface again, like herpes--or China.

Not going to disagree with that sentiment
The market will probably react immediately once someone abuses that power
Anyways been hearing of ways to fix this problem but this is a puzzle for developers and the networks to find a solution to

Does anyone have data on double spends that go back further than this?: https://blockchain.info/double-spends

Thanks for this and for the links that explain how 51% isn't automagically a disaster.

Does anyone have a link or explanation for how Ghash.io became so popular?  What is so great about them?  I've been to their site and don't see anything super special or am too ignorant to recognize their awesomeness.  Their rise, regardless of what they do with it, seems pretty remarkable.

Thanks!

For cex site, people pay to get hashing power. Cex uses a whole bunch of miners for the ghash pool. The fees are high but I guess people like trading the hashing power to the next person (for a quick profit).

There are probably two reasons why cex/ghash is so popular

One - they are very good at marketing and making their website look good. If you look at their stats pages, their worker pages, the trading platform, the balance history it is very organized looks neat, is very detailed and is appealing to the eye. Now look at the stats page for eligius, the stats are nowhere near as detailed, you don't have as much information and someone who does not have a lot of knowledge about bitcoin may not be able to interpret what everything means. I think eligius is a very good pool and that absolutely nothing is wrong with it, and you generally will make exactly the same with both pools (they use different payment methods PPLNS vs CPPSRB [capped paid per share with recent backpay]).

Two - They offer merged mining with several altcoins. This will increase profitability by a little bit (the small amounts matter in mining) and offer a platform to sell two of the altcoins that are merged mined.

I just find the topic odd simply because it would be a bad investment the 51% attack took place, so why would people put their investment at risk by not simply mining elsewhere?  Probably something I've overlooked but hey I don't mine BTC in the first place.

Even though ghash is now at 30% I don't think this question is answered at all and this still very much an issue. This is because an attacker could could potentially gain control over more than one pool or more than one pool could collaborate to execute an attack. I would really like to see the pools such that not even any given 5 pools total to 51%. Ideally we get rid of centralized pools alltogether or some serious protocol change is made to make it easier to solo mine, or so that 51% attack doesn't work any more alltogether.

It isn't a protocol issue, it is more a human behavior issue and that is a lot harder to "fix".   Logically a miners largest cost is the electrical bill and electrical bills are only due once a month.  Even a pool with 2% of the hashrate will have a 95% confidence to be +/- 10% of the expected revenue after 30 days.   However most miners want a reward now, right now.  That probably isn't going to change. p2pool is a good example.  Right now p2pool shares are ~1000x easier than solo mining however most miners do use it.  Logically even if you radically changed the protocol such that solo mining was 1000x easier most miners would probably not use it. 

Still the risk of 51% attack from pools is overstated.  Lets say right now ghash had 70% of the hashrate.  How would they produce an attack chain?  Well they would need to stop mining the main chain.  ghash on average would go from producing say 8 blocks in a 2 hour period to producing zero.  They can't mine both the main chain and attack chain at the same time.

The risk is small but it could be made even smaller by making it easy for miners to verify they are extending the longest chain.  For example the current block hash of the last block in the longest chain is 0000000000000000302b107dcfdaf35d3b09366638e19aa24ffa91dd7f91d57c and thus all miners should currently be working on a block that extends that block.  Right now if all pool miners are extending block 0000000000000000302b107dcfdaf35d3b09366638e19aa24ffa91dd7f91d57c then you know that no 51% attack is ongoing.  Even if all pools were infiltrated the only thing an attacker could do is steal the mining revenue (by changing the coinbase tx). 

It would be possible to modify the blockheader so that miners (and third party observers) can verify that they are mining the longest legit chain.  This could be done in a backwards compatible manner (won't break existing ASIC hardware) by including a partial block hash of the prior block in the last segment of the block header.  There is limited space in the final segment of the blockheader (the one that the miner adds the nonce and thus must always be provided in full by the pool) so the full blockheader is far too long.  By extending the blockheader and putting the rightmost 8 bytes of the prior blockhash (i.e. 4ffa91dd7f91d57c for the block above) in the last segment it would be easy to spot attempted misuse of miner computing power.  The mining software (using SPV protocol) would maintain an independent record of the longest chain (and the last block extending it) and would verify the work provided by the pool, extends that chain.  If the software detects a scenario where they are not mining the longest chain they would stop mining (or switch pools to a pool that is extending the longest block) and warn the user.   

This type of "verify I am on the longest chain" can be done without a change in the blockheader but it would require pools to send the full blockheader to miners.  For efficiency reasons they don't and that is unlikely to change.  Also if a pool doesn't it can't be forced to do so and most miners wouldn't see a lack of the verifying information as sufficient reason to not use the pool.  Changing the blockheader would ensure the necessary information is always sent to the miner.

no, you're completely right. they wouldn't do it. it would be pointless, unless your point was to destroy bitcoin and the wealth of all the miners involved. the retards like failing seem to think some 'organization' is going to 'take control' of ghash and essentially perform a very expensive terrorist attack on bitcoin. its not to make money, even though the only reason they've cited has been 'greed'. but those guys are VERY dumb so i wouldn't take much of what they say into consideration.

Also, even if they did it and all PoW coins crashed to 0$. Peercoin would still remain working as if nothing happened. Thus, they cannot destroy crypto currencies as a concept by performing a 51% attack. All sane bitcoin investors have diversified between alts long time ago, so even their wealth cannot be damaged by such an attack. All in all, 51% attack is not a realistic threat. At least not for the (s)perma-bulls.

These guys seem to think we are doomed:

http://hackingdistributed.com/2014/06/13/time-for-a-hard-bitcoin-fork/

http://hackingdistributed.com/2014/06/16/how-a-mining-monopoly-can-attack-bitcoin/

While I was somewhat comforted by the fact that a 51% attacker can't take my coins, the fact that they can basically destroy the other pools is a bit alarming.  They also do a good job talking about what Death and Taxes was referring to as the human problem.

I just don't see what keeps other pools from being as attractive as Ghash.io, that would help with the issue.

GHash is currently 44% ... What the fuck people >:(

I see 40%
https://blockchain.info/pools?timespan=24hrs

I guess the market forgets after a while short term attention span and or it gave the community time to make a solution and seeing none is just going back to its old routes.

What changes would you propose to implement to make solo mining easier?

The Bitcoin difficulty is based on the network hashrate and the network has a hashrate that makes it so most small to medium sized miners would not likely find any blocks during any two week period. Pool mining is the solution to this problem.

I have no particulator solution i.e. I'm providing requirements but not analysis/design.

you guys figured it out yet ?
report back when figured... if problem.. fix... if not... report back amusing story

thanks :)
so many threads about this...

51% attack vulnerability is just one of bitcoins core flaws that won't be fixed. hybrid or pure PoS mining is adressing this issue.

Even if the people who run the pool, stay nice, then there can always be security vulnerabilities that will give access to people who don't have plans to stay nice.
And 51% attack isn't only a threat if a single pool reaches it. If already 2 pools make up 51%, and both of their security is compromised, then BTC is still screwed.