Bitcoin minting is thermodynamically perverse

Let me begin by saying that Bitcoin is an amazing project and I am very impressed with the implementation and the goals. From reading these forums it seems to be understood that debate about the design and operation of the bitcoin economy ultimately serves to strengthen it, so I hope these comments are taken in that spirit. *EDIT - I have been convinced by further research and discussion that Bitcoin is actually highly efficient compared to most traditional currencies, because the infrastructure required to support a government issued fiat currency represents a much larger investment of resources than Bitcoin's cpu power consumption. I am leaving this thread active though because it has been generating a lot of interesting discussion.*

I believe that the amount of energy input required to the bitcoin economy represents a serious obstacle to its growth. I think in the long-term, transactions may be even more serious than minting in this regard, but I will for the moment discuss minting because it is more precisely bounded and defined. The idea that the value of bitcoins is in some way related to the value of the electricity required, on average, to mint a winning block is generally accepted, but the precise nature of this relationship is contentious.

One argument is that anyone who chooses to generate coins is actually making the choice to purchase bitcoins with electricity/computational resources, and that because some/many people are in fact making that choice, bitcoins have at least that much "value" to the generators, who can be assumed to be maximizing their utility. A contrasting argument is that cost of production is different than market value, and the most objective measure is the current market conversion price to a more liquid and widely traded currency such as the US dollar.

My contention is that both of these arguments miss the point and the real problem, which is the fundamental perversity of wasting large amounts of energy and computations in generating the winning blocks for the minting process. The minting process exists because of the necessity of actually "printing" the currency, and certain desirable properties of crypto-math for making the currency's behavior predictable. The fact that the current minting process requires a large energy input of computational work is highly unfortunate and has the perverse consequence that bitcoin may actually be "destroying wealth" in the sense of wasting energy producing a digital object worth less than the resources invested in it.

As is often pointed out, a currency does not necessarily have, or need to have, any inherent value - a medium of exchange is a useful tool and can have value purely as a consequence of social convention. The cost of production of bitcoins in electricity consumed represents a waste, a "thermodynamic burden" that the currency has to carry. Consider a hypothetical alternative digital currency called "compucoin", which purchases cpu cycles from nodes on the network. The market value of this currency would converge very closely with the cost of electricity required to generate cpu cycles. Instead of costing cpu cycles to mint, the value of the cpu cycles the coins could be exchanged for would create a rational basis for the currency's value and integrate it with an existing market. I imagine that alternatives to Bitcoin (many of them probably sharing a lot of Bitcoin's source code) will inevitably emerge and Bitcoin's current minting process makes the currency "expensive" in terms of energy input. I believe this places it at a competitive disadvantage to other currencies and can only hinder its widespread adoption and long-term value. *Edit - as mentioned above, I am now much more optimistic about Bitcoin long term. I still think compucoins would be a cool idea, though!*

Thanks very much for the extensive and informative reply. I do not disagree with the points that you made, but I also don't believe it invalidates my fundamental point: it is not inherently necessary for a digital currency such as bitcoin to require as much energy input as bitcoin does. A digital currency that offers bitcoin's behavior at a "lower cost" of energy overhead of the currency and transaction system has a competitive advantage. The digital currencies that will win and become standards, I believe, are the ones that offer the most value-added on top of the costs of running the system. Of course, there is room for many currencies - we already have a multiplicity - so it is not necessary for Bitcoin to become the "one true money" for it to succeed, but I believe the current minting policies will be harmful to its growth and adoption long-term. Much of the utility value of bitcoin resides in the work done to establish the security and reliability of the system - accomplishing that work with a smaller energy input would seem to be beneficial.

Nicely written gridecon. I completely agree.

I've been considering posting a similar critique but you've laid the issue out clearly. if you don't mind I'll add my thoughts here.

As the minting process and the transaction recording process are one, there is really no reason to separate them. I don't disagree with knightmb's points at all. But I do think that is is really important to point out that all that CPU power and electricity usage is absolutely NOT required for the task at hand.

Generating blocks serves three critical but independent functions in the bitcoin system.

1. It permanently records valid transactions in a roughly chronological order.
2. It creates *consensus* among all nodes on what transactions took place and when, WITHOUT relying on a central authority. That was a hugely clever breakthrough.
3. It trickles bitcoins into circulation at a regularly scheduled rate. (roughtly 50 BTC/10 min) Also a nice motivational trick.


However, all of these required tasks could be done in a much more efficient manner.

Generally in a P2P system, if you add more nodes then the necessary machine resources needed per node decrease. (for a given usage level of the service)
However, if we add more nodes to the bitcoin network the necessary machine resources needed per node increase. This is true of CPU usage and network bandwidth.

Satoshi already pointed out that the goal of this is not to scale to millions of pure P2P nodes. But instead to have thousands of transaction checking peers each having thousands of clients. That means the goal is to create core "central authority" of peer nodes each of which is trusted by some clients, but the core peers do not necessarily need to trust each other. This is a absolutely critical design construct if the current implementation of bitcoin is expected to scale to millions of traders.

However, there are other design possibilities that could meet each of these three critical goals without exponential growth in resources utilization.

Implementing the transaction graph using a redundant distributed hash table comes to mind. This could be done with or without a core of central peers. Another possibility would be just to store the transaction graph in a SQL database and use database replication among the core peers. Either of those satisfies goal 1 and 2 using minimal system resources.

Goal 3 is implementable by simply generating a distributed shared secret, then using that shared secret as a concensus random node to send 50 BTC to.


I agree with gridecon that such an implementation would out compete the current implementation. Even if it otherwise following exactly the same external behavior and transaction validation rules.

Go for it.  In the meantime, I think that you both miss some fundamental truths.

1)  The point of the system is not to 'mint' coins at a high energy cost, but a high computational cost, to protect the system no matter how high Moore's law may go. 

2)  The award of bitcoins is not 'minting' really anyway, it is an award by lottery for participating in the strenghtening the currency.  Those 50 bitcoins already existed from the moment that the rules were conceived and the process set into motion.  They are just distrubuted from the original owner in a systematic manner.

 The computations are a neccessary part of keeping the system strong and confidence high, and are a small price to pay when compared to the costs that we all incur from the administration of fiat currencies.  I think that it was a brilliant idea.

I understand the bitcoin is not INTENDED to be about the minting. The essence of my criticism is that a digital currency SHOULD NOT be "about the minting" but the current bitcoin system works against that goal! My argument is that the current minting system limits the utility of the currency largely because entirely too much attention is focused on "winning the block generation lottery" and the process of choosing these winners is highly inefficient and resource intensive. If potential users of the currency believe that bitcoin is simply a way for botnet herders to convert spare cpu cycles into cash, they are unlikely to participate in the emergence of the economy. The current system could easily result in the perception that bitcoin is basically being used as a vector for the indirect monetization of the theft of electricity via compromised computer systems. I assume most people interested in bitcoin understand that self-interested individuals will obviously seek to exploit opportunities like this, and the design of a popular currency has to prioritize how a prospective user will evaluate its fairness.

Thanks to everyone for all the well-considered responses. I am, you could say, half-convinced - I understand the motivations for the design decisions bitcoin has made, but I still think more optimal solutions are likely to evolve. That said, the obvious intelligence and open-mindedness shown by the discussion here is in some ways the most convincing selling point! Any currency with a user community who engages in quality dialogue earns additional credibility with me, so based on the quality of this thread, I have acquired a few bitcoins, and distributed some of them to the posters who included a bitcoin address. As I hope I have emphasized, I think bitcoin is brilliant and I want it to succeed, I raise these issues because I would hate to see the success of bitcoin impeded by something, the minting model, that everyone agrees is in many ways arbitrary.

I think that's the issue you have though, you think it's inefficient and resource intensive on purpose when really it's not. No one is required to generate coins to use the system. If you click the generate coin list, you are basically volunteering your resources to strengthen the network and in exchange, you *might* get some coin for it. Since coin is in demand to use the system, someone has to generate it. If generation was as easy as "input how many you want, *bam*, you have 1 million now" it would make the entire system pointless. It's a chicken and the egg puzzle except there really isn't a puzzle here. People need bitcoins, they generate them at a fixed rate to make it worth spending the CPU time.
There's no reason to think of it that way, anymore than the local barber is collecting hair to clone an army of super-men for world domination. You have to draw the line somewhere. If someone is going to steal electricity, targeting a 30 watt computer doesn't make a lot of sense.

I have seen the topic come up here a few times "what if someone seeks to exploit this?" and the answer is the same over and over, it has been designed from the ground up to counter this exact concern. If you want to "game" the system, you need a lot of CPU power. CPU power cost money. The system scales itself to the available CPU power it has so if you bought 10,000 EC2 sessions from Amazon and started a massive bitcoin generating farm, you would never make more money out of it than what you put into it for long because the system would self-adjust against it. If someone throughs a massive botnet at it, the same thing will happen. The system is set for self-feedback, so to try to steal from everyone is to try to steal from yourself at the same time.

The computational and energy cost for generating bitcoins is quite irrelevant to the bitcoin economy since it does not compromise the integrity of bitcoins. People should be more concerned about a botnet trying to cheat rather than mine bitcoins.

You seem to have things backwards.
The "dice rolling" is to maintain the "chain of proof".
Actually giving out the coins is a side effect.

I have 2 questions for you:

Without high computational burden how do you propose we:

a) Maintain the integrity of the system?
b) Distribute/Seed the system with coins?

You mean it is designed  to "trust no one", not of "trust everyone collectively"?

This exchange is fascinating! At some sense it's just a matter of semantics, but it is very insightful semantics.

Logically I can support knightmb's statement "trust everyone collectively".
However kiba's statement "trust no one" is disprovable,
as is knightmb's statement "Trust no one node, but if enough nodes are saying the same thing, then trust that."

The way the system works is every node attempts to receive every transaction, they each try to put every received transaction into a block at the same time. And while they are doing that, they each play the hashing game individually looking for fate to bless them with a match before it does anyone else.

So at any given point in time the system is really *trust (but validate) someone arbitrarily*. This happens when someone declares a block & hash "solved". All the other nodes check the block for self consistency, but they DON'T check the block's transactions for consistency with the transactions they themselves were working on. Nor do they check against the transactions other nodes received. So I see no support for the "if enough nodes are saying the same thing, then trust that."

All nodes simply declare, "I will trust this node's work as the next increment of bitcoin truth, because fate has blessed it." So it doesn't qualify as a "trust no one" system. Node's never say, "Hey you fuck head! I have 5 known transactions you didn't bother to put in your block! You are a lying bastard." Everyone just presumes the missing transactions will be resubmitted and they'll end up in a subsequent block. This is heuristically plausible, but not absolutely deterministic. Everyone has to keep checking to make sure their transactions actually cleared the process.

So if knight's "trust everyone collectively" meant, "trust in the process" I find that insightful and a good motto. I'd probably add "...but keep doing your own validation!"

Ah I understand then, the answer is botnets can't forge currency because the swarm would not accept it. The only thing they can do is produce "real" currency at the expense of the stolen CPU time. Just like forcing slaves to mine gold will still net you real gold, just at no cost.

The only known *practical* attack to bitcoin from a botnet would be to have more CPU than the swarm at your command and to use that CPU power to double-speed currency. So while that could possible fall in into the "forged currency" category, it's not so much to make fake currency as it is spending valid currency twice before anyone notices.

Ok, bravo for an amazingly honest and straightforward answer. That is probably the most overall convincing argument I have seen made in the thread. You are arguing that the overall social utility of Bitcoin is "worth it" - worth both the raw energy cost of the computational work, and worth the regrettable fact that it provides a profit opportunity for botnet operators. I don't have a proposal for how exactly you measure the costs and benefits, but I entirely agree with your analysis of the bottom line of the situation. You could even make the argument that minting bitcoins as an activity is less harmful than other botnet activities like sending out spam.

Well said.

Participation in the network as an honest node helps everyone.

It's the same situation as gold and gold mining.  The marginal cost of gold mining tends to stay near the price of gold.  Gold mining is a waste, but that waste is far less than the utility of having gold available as a medium of exchange.

I think the case will be the same for Bitcoin.  The utility of the exchanges made possible by Bitcoin will far exceed the cost of electricity used.  Therefore, not having Bitcoin would be the net waste.

Each node's influence on the network is proportional to its CPU power.  The only way to show the network how much CPU power you have is to actually use it.

If there's something else each person has a finite amount of that we could count for one-person-one-vote, I can't think of it.  IP addresses... much easier to get lots of them than CPUs.

I suppose it might be possible to measure CPU power at certain times.  For instance, if the CPU power challenge was only run for an average of 1 minute every 10 minutes.  You could still prove your total power at given times without running it all the time.  I'm not sure how that could be implemented though.  There's no way for a node that wasn't present at the time to know that a past chain was actually generated in a duty cycle with 9 minute breaks, not back to back.

Proof-of-work has the nice property that it can be relayed through untrusted middlemen.  We don't have to worry about a chain of custody of communication.  It doesn't matter who tells you a longest chain, the proof-of-work speaks for itself.

I can't imagine a more splintered network.

You would then have to rely on "for profit" companies to be the central authority for the new currency and how long would it take to generate the full chain even at a difficulty of 1.00?

Thanks very much for your reply. I agree with your analysis, and this thread has actually changed my mind as to my initial criticism. After more careful study of the design of the Bitcoin network and trying to understand the exact manner in which Bitcoin attempts to create value from the computational work invested, I am now inclined to think that bitcoin is in fact high EFFICIENT rather than inefficient. My thinking now is that bitcoin does not, in fact "waste" computational work at all - instead it works hard to deliver the most value possible from that computational work. Something like a governnment issued fiat currency may not have any obvious energy burden beyond its printing - but in fact, maintaining the value of a fiat currency requires a substantial investment in maintaining police enforcement, a legal system, and national defense. In comparison to the energy cost of hiring police officers to enforce economic honesty, the energy costs of investing cpu cycles in guaranteeing that honesty mathematically seem very small!

I really I understood the point of this thread but I guess I didn't.

I assumed that if you could design to do the exact same thing in the same commodity quantities and at the same protection level, BUT consuming less energy and producing less BTUs of heat, then that would be less wasteful.

You can mine gold in lots of ways, some require less resources than others. If two processes produce the same amount of gold then no point in optimizing anything else?

The heat from your computer is not wasted if you need to heat your home.  If you're using electric heat where you live, then your computer's heat isn't a waste.  It's equal cost if you generate the heat with your computer.

If you have other cheaper heating than electric, then the waste is only the difference in cost.

If it's summer and you're using A/C, then it's twice.

Bitcoin generation should end up where it's cheapest.  Maybe that will be in cold climates where there's electric heat, where it would be essentially free.