Topic: Is the current level of computational power necessary to secure the network? (Read 1179 times)

It's actually not high enough since the security of the network is presently under constant threat by Deepbit, and has been for at least a month by now.

The steady state of Bitcoin is when new coin generation is negligible and mining is rewarded with transaction fees. Currently the amount of mining is indeed too high for the network security requirement, since it actually serves 3 purposes:
- Having an objective way to make the initial distribution of coins
- Securing the network
- Practicing mining for when it will really be needed to secure the network

Going forward, there will be a tradeoff between the inconvenience of paying fees and the need to incentivize mining to secure the network. There's been discussions of how to reach a good tradeoff but nobody knows what exactly the future holds.

A lot of people want to look at two factors and try to make a decision about which one is first, and which one is second.  Things are rarely that simple and direct.

Bitcoins are valuable because the network is secure, and the network is secure because bitcoins are valuable (that is, because there is an incentive to increase security).  The two chase each other endlessly and cause each other.  You can't pick one and say that it is the cause and the other is the effect.

If you are thinking in terms of the safe analogy (which isn't really very good), remember that everyone in the world is free to use this safe.  Having it be far more secure than it needs to be will invite more people to use it, which makes the safe itself more valuable, which makes the safe more secure, which invites more people to use it...

And since the network is distributed, no one person gets to decide how much security is appropriate.  If you think that $4 million per year is too expensive to store $10 million, you are wrong, because the rest of the planet has voted, and the consensus is that the cost is exactly right.  (Yes, I know, made up numbers.  Think of them as references to the current cost and current value, whatever they happen to be at any given moment.)

That is exactly what difficulty is, a worldwide consensus about how much effort we are collectively willing to spend to protect what we have.

A huge BTC market cap could a have small number of only low value transactions requiring very little network security. Similarly, a small BTC market cap could have a large number of high value transactions requiring a lot of network security. So I still don't think the network security needs have a lot to do with BTC price, but I do see your point that the network security needs have a lot with how enticing it would be to steal by reverse&respend. I am now thinking that the network security needs are most closely related to the value of transactions being performed. Anyway, I still don't see how how the supply of computational power (i.e. BTC price) always happens to exactly balance out the demand for computational power (i.e. value of transactions).



I see your point about more security being good, but another way of looking at things is to question whether it makes sense to pay $4,000,000 per year for an ultra secure safe to store $10,000,000. In this example it might make sense to “have a weaker lock”.

If you view the Bitcoins being awarded to miners as a network security costs, you can calculate the percent of the Bitcoin market cap that goes to pay for network security. For example, currently bitcoins are being awarded to miners at a rate of 10,500,000 bitcoins for the first four years (i.e. 2,625,000 per year) and the current total number of bitcoins is 6,615,750. This seems like a huge portion of the market cap going towards network security and so I wonder if it is really necessary. It just seems like a lot to spend 2,625,000 BTC in a year on network security especially since I don't believe there has been one report of a reverse&respend.

This is like asking if your safe could have a weaker lock. Probably yes, but lets not.

No, it doesn't have anything to do with the _number_ of transactions.  The effort required to secure the network comes from how beneficial reverse&respend and transaction denial of service of service attacks are.

So this is related to the market cap of bitcoin and how easily someone can make irreversible transactions using bitcoin.

Unfortunately since we have single parties bumping 40% hashpower at times (and pairs of parties >50%) we're not actually doing very well on the security front right now.  It would be hard for an outsider to attack the network, but by compromising someone in control of a lot of hashpower attacks could be accomplished easily within the system.

When I read the Bitcoin wiki FAQ it says, “The computations done when mining are internal to Bitcoin and not related to any other distributed computing projects. They serve the purpose of securing the Bitcoin network, which is useful.” That quote leads me to believe that mining computations are only used for securing the network, but it leaves me wondering if all the computations are necessary.

From my understanding, the supply of mining computational power has a lot to do with BTC price whereas the demand for mining computational power has a lot to do with the number of transactions being made. I just don't see how the supply of computational power always happens to exactly balance out the demand for computational power.  The collective supply of computational power provided by miners has gone up by orders of magnitude over the past few months, but I don't see how the demand for computational power could have possibly increased proportionally. Is the over-abundance of computational power being used to secure the network by rechecking things many times over? Is there really a need for the current high levels of computational power to keep the Bitcoin network secure?