Newb Question on Security

[CautiousFan]

I just started following the Bitcoin world about a week ago.  I've entered into a lively debate with a coworker about its viability and was hoping someone might shed some light.  Here's my concern.

If the nodes are all rational, then they'll only perform work verifying transactions if the verification fee is sufficient to cover costs.  But a nation-state might be willing to perform a tremendous amount of work beyond the threshold where it ceases to become economically viable (and do it with economies of scale too).  And the nation-state doesn't have to compete with the total POTENTIAL CPU power of the network.  Because in reality, the potential size will never exist, but only what is economically viable or, to say another way, what the fee structure can justify sustaining. 

So lets say the total monthly fee's total up to equivalent 1 billion dollars.  A rational network performing at marginal cost will consume 1 billion dollars in resources to earn the fees.  No more.  If bitcoin becomes a threat to "national security" the U.S. might spend 2 billion dollars a month (pocket change) to destabilize the system, and they may get back 1 billion dollars of that.

If this problem is real, I can think of 3 different solutions, none of which I like.  I think the best way to solve this problem is to raise the transaction fee, thereby garnering more resources.  At the same time, you've got to make it impossible for someone who's attempting to game the system to get paid...a difficult task by definition.  Other solutions are 1) relying on good will or 2)figuring out a way to force participants to perform lots of verification work.

So am I in left-field with this?  I hope so.  Please help me resolve my debate. 

[1714000094]

1714000094

[1714000094]

1714000094

[1714000094]

1714000094

[1714000094]

1714000094

[1714000094]

1714000094

[Insti]

Not all the nodes will be 'rational'

[RHorning]

The network is built on a nearly complete lack of trust in other nodes, which is sort of the point.  Yes, it is possible that somebody like the NSA could have some top tier programmers really push for some research to destabilize Bitcoins for a political purpose (presuming those computer scientists at the NSA has nothing better to do with their time).  Certainly if there are billions of dollars equivalent of bitcoins going through the network (even on an annual basis) there would be some people motivated to try and "skim" a few more for themselves beyond transaction fees and so forth.

What is the saving grace here is that this network does not rely upon "security through obscurity", but rather mathematical relationships that are apparent to those at least intelligent enough to understand the principles involved.  As the open source saying goes, "with many eyes, all bugs are shallow".  As can be demonstrated in regards to the security for the domain name system of the internet, it is possible for something widely used but discovered to have a major vulnerability to be fixed, and fixed "quickly" even if there are flaws that are later discovered after deployment.  Indeed I would suggest that now is the time to find the flaws in this system, and really poke at it hard to see if you can come up with something that would break the network.

The longer that bitcoins is working, the stronger it can rationally be said to be, particularly as it moves from hundreds to thousands of users now with the latest influx of users.

I would like to have some more information about the transaction fees and how they are throttled in terms of somebody processing a transaction and taking 90% of the bitcoins in the transaction as "fees".  The transaction processing is built into the network, however, and something that has been already planned. 

It should also be noted that if the network is processing monthly fees of the equivalent of a billion dollars per month, that billion dollars isn't going to be concentrated in the hands of just a few people.  It would be disbursed among millions or even a billion different people so that the amount earned for a single person would still be a buck or two per month.  Perhaps a bit more or less depending on how heavily used the system will be and your network bandwidth, but it would still not be concentrated in the hands of just a few people. 

Where a government might get involved is to "force" the network through just a few "gateway" servers for international commerce, earning those governments some extra money on the side.  I'm thinking the Great Firewall of China here or something equivalent, and that could potentially be implemented so far as changing the raw topologies of the internet as a whole to force it to happen.  Even that has a few work-arounds for the really paranoid, such as some of the messaging with FreeNet and some of the other paranoid peer-to-peer networks.  Messages would route around censorship and manipulation as a form of damage in the network.

[CautiousFan]

Insti #2

I agree with you.  My account isn't rational cause I have philosophical reasons for wanting this to work.  BUT, if you really want this thing to spread, it's got to be rational for people to participate. 

RHorning #3

I reread your post several times, but I'm not seeing an answer.  Let me try and repost the problem I see.  As I understand, Bitcoin works because it relies on the combined CPU power of the entire network.  But it is not rational for everyone to give all their excess CPU power, and trying to force people to do this via programming changes (mentioned in other posts) will encourage the development of cheats.  But if everyone only gives a rational amount of CPU power (even adding a few thousands fanatics), then the system can be defeated by a gov't which has more to loose then the total system fees.   A system likes this strikes at the very foundations of statist power...there is a lot to loose.  I don't think they even require crazy quantum computers or secret code hacks.  Just more power.  Even if a billion people adopt this, states may pool their resources to outmuscle it.  It just hurts them too badly to ignore it.

But you know, I'm really not smart enough to really understand this thing.  I'm probably just blowing blabbering about something trivial.  But I do know, if this thing works.....hold on to your hats.  It'll be HUGE.  A geo-political meat grinder like the world has never seen.   

[CautiousFan]

There's another thread that started about the energy wastage that occurs in Bitcoin.  This is what I'm getting at.  I'm not sure a secure system is economically rational for it's participants, and this could be an achilles heal. 

[Insti]

I reread your post several times, but I'm not seeing an answer.  Let me try and repost the problem I see.  As I understand, Bitcoin works because it relies on the combined CPU power of the entire network.  But it is not rational for everyone to give all their excess CPU power, and trying to force people to do this via programming changes (mentioned in other posts) will encourage the development of cheats.  But if everyone only gives a rational amount of CPU power (even adding a few thousands fanatics), then the system can be defeated by a gov't which has more to loose then the total system fees.   A system likes this strikes at the very foundations of statist power...there is a lot to loose.  I don't think they even require crazy quantum computers or secret code hacks.  Just more power.  Even if a billion people adopt this, states may pool their resources to outmuscle it.  It just hurts them too badly to ignore it.

But you know, I'm really not smart enough to really understand this thing.  I'm probably just blowing blabbering about something trivial.  But I do know, if this thing works.....hold on to your hats.  It'll be HUGE.  A geo-political meat grinder like the world has never seen.   

1) I suspect you are missing some factors from your 'rationatlity' assessment.

If I have a large stock of bitcoins, I want to keep the system alive so that I don't lose my value. In this case it is rational for me to expend more in energy costs than I would be rewarded for block processing.

I agree bitcoin generation should stabilize at the energy breakeven point, with processing power coming on and offline in line with market movements.
But, by the time this happens the amount of processing power involved will be HUGE and I suspect no government will have enough to take it offline by technical means.

2) How will a government destroy the system with excess CPU power?

[InterArmaEnimSil]

I just started following the Bitcoin world about a week ago.  I've entered into a lively debate with a coworker about its viability and was hoping someone might shed some light.  Here's my concern.

If the nodes are all rational, then they'll only perform work verifying transactions if the verification fee is sufficient to cover costs.  But a nation-state might be willing to perform a tremendous amount of work beyond the threshold where it ceases to become economically viable (and do it with economies of scale too).  And the nation-state doesn't have to compete with the total POTENTIAL CPU power of the network.  Because in reality, the potential size will never exist, but only what is economically viable or, to say another way, what the fee structure can justify sustaining. 

So lets say the total monthly fee's total up to equivalent 1 billion dollars.  A rational network performing at marginal cost will consume 1 billion dollars in resources to earn the fees.  No more.  If bitcoin becomes a threat to "national security" the U.S. might...
So am I in left-field with this?  I hope so.  Please help me resolve my debate. 

You're quite correct in your assumptions, but the debate is moot.  If bitcoin becomes a threat to US security, they'll make it illegal, or come after its top users legally, or shoot people.  No need to go into theoretics about how to protect the network when you're talking about competing against a nation with millions of armed troops and nuclear weapons.  Unless you're the premier of China, you can't compete.

In reality, if bitcoin becomes a big deal, it *will* be government-regulated, whether we like it or not.

[CautiousFan]

Insti

You know, I'm just some dumb schmuck with a keyboard and I seriously don't know what I'm talking about so let me just refer to some other sources who may actually have credibility. 

Wikipedia states "This mechanism is claimed[1]  to be virtually tamper-proof. For an attacker to manipulate the record, he must outpace all of the other nodes on the network to produce the longest proof-of-work. This becomes exponentially more difficult as time passes (I assume because the system gets larger)."  Satoshi's white paper article considers what happens when an economically rational attacker (the greedy merchant) does this (section 6). However, he does not seem to address the threat from those with much larger economic incentives such as a nation-state, whose goal might be to destroy trust in the entire system.

I don't know how tampering occurs, my point is that if the system is using it's resources rationally (including vested interests you mentioned), a nation-state (or a group of them) have sufficient incentive to amass the CPU power to outpace all the other nodes in providing the proof-of-work calculation.  Once they have pole position, my impression is they can somehow use that position to harm the system. 

[ichi]

You're quite correct in your assumptions, but the debate is moot.  If bitcoin becomes a threat to US security, they'll make it illegal, or come after its top users legally, or shoot people.  No need to go into theoretics about how to protect the network when you're talking about competing against a nation with millions of armed troops and nuclear weapons.  Unless you're the premier of China, you can't compete.

In reality, if bitcoin becomes a big deal, it *will* be government-regulated, whether we like it or not.

I agree that governments will seek to regulate and police the system, if/when it starts to move substantive value.  The US, for example, will undoubtedly label those who evade its control as criminals and/or terrorists.  Some governments will shield domestic corporations, and those who do business with them, just as they do now.  However, as USB depositors have found, such protection may be ephemeral.

Those who reject government control would be wise to learn the skills necessary to ensure their safety, and the safety of their associates.  Government control is not inevitable.  Armed troops and nukes are useless against distributed systems of anonymous participants.

Edit: I just came across this.

[FreeMoney]

Thanks for those links ichi. I haven't finished the gun one yet. Both are very cool.