Topic: Bitcoin & Tragedy of the Commons - page 4. (Read 21903 times)

To put it simply what will happen is this:

We will have a large number of pools and most likely many different p2pools as well. I will pull some numbers out of my ass to illustrate this, in reality the market will decide these numbers.

Pool x has a fee threshold of 0.1, meaning that they only accept transactions that have at least 0.1 bitcoins as a fee. This is the "high speed, high prices, low volumes" pool.

Pool y has a fee threshold of 0.05, which is "if you want decent speed but don't want to pay a premium" type pool.

Pool z has a fee threshold of 0.001, this is the "cheaper than the competitors, but slow as fuck" type pool.

These are just a few examples, in reality there will be a large number of different fee thresholds and there might even be pools that accept basically everything. These could be a "donate your computing power to the Bitcoin project" type pool.

From the user's perspective this will look like this:

Bitcoin client tells a user that at this moment in time, the average speed for a transaction with a 0.1 fee is 10 minutes (because in this example this is the fee that basically everyone accepts). With a 0.05 fee you'll get it through in 60 minutes on average. With a 0.001 fee it will go through in 10 hours. A fee of 0.0005 is not recommended but it will go through in a day or two.

Now keep in mind that the actual numbers are meaningless here, they are just an example. Fees valued in bitcoins could actually go down significantly if Bitcoin value skyrockets but one thing is for sure, fees will go up significantly when valued in dollars/euros. It is difficult to predict how much higher the fees will be but most certainly lower than any competitive technology or method of money transfer that has been developed so far.

This will actually form a brilliant market where miners will change which pool they support because they are constantly tracking which transaction fee class is most profitable. Yet at the same time it will vary how much users are willing to pay for certain speeds and if too many miners switch too fast to a lower threshold it can suddenly become unprofitable for them because a higher class is forced to lower their threshold and thus a lot of users will switch to faster transactions because they now seem cheaper.

I could go on for much more on this but this is good enough for now.

I propose that somebody go and do the following, which should increase the security of the network regardless of whether we have a "tragedy of the commons" situation.  This is essentially a boil-down of something proposed in the Stanford paper, and something I have proposed in the past.

• Create a patched Bitcoin node that automatically rejects (ignores) any reorganization of 6 blocks or more.  (To use the Stanford paper terminology, this means automatically checkpointing any block that's 6 blocks old.)
• Run it on good hosting resources and have it accept incoming connections, and have it connect to a larger-than-usual number of nodes, so as to maximize the likelihood that it will be on the "good" (popular) side of any major blockchain conflict.

Why?  If and when we ever get attacked by a replacement chain, a few trusted people running nodes like this will save the day.  The Bitcoin community will say "oh shit", but soon, word will get out that the recovery procedure (for the typical user) is as follows:

• Delete your block chain
• Start up your bitcoind, using the -connect parameter to only allow connections to these specific nodes that ignored the unwanted blocks, and refuse all incoming connections.
• Enjoy business as usual, while the devs and everyone else dance around and figure out how to go back to a decentralized system that also excludes the efforts of the attacker.

What this amounts to is an increased form of democratic trust.  Yes, I understand it is centralized - its purpose is temporary.  Proof of work is the primary decider of valid blocks, but in the case where the community at large sees that the blockchain has been thwarted in an obvious manner, users reserve the right to "vote" for some particular view of the block chain by connecting only to nodes known to be espousing it, without having to worry about modifying their clients beyond their abilities.

A miner can try to demand high fees, but if 99% of miners accept low fees, nobody will be willing to pay these high fees.

You are forgetting that (unless limitations are placed), space for transactions is an abundant resource. It does not work like a market where participants manufacture a scarce product. The only thing to lose by including a transaction is bargaining power - which is everyone's, not just the individual miner's.

That's exactly the problem, miners will quit and the network will be less secure.

I don't think you quite grasp the way markets work. What you're describing makes no sense to me. The mining network will form a near-perfect market environment where some will have a business based on volume and some will base their business on higher prices and faster transactions. Miners won't just quit in the same way as they do now, if the fee they are accepting suddenly becomes unprofitable, they can simply change their fee threshold to find a more optimal level. A more optimal level could just as well be higher, not necessarily lower.

If the mining community is too crowded compared to Bitcoin transaction counts (and to an extent price and difficulty as well, just like now), then some will quit mining and thus difficulty goes down meaning more blocks and thus more reward for everyone else (not as block rewards, but as transaction fees).

The only real change this involves for Bitcoin users is that transactions will not be as cheap as they are now. Other than that Bitcoin will be much safer than it is now (assuming transaction count keeps growing long term). It will require some changes to the way bitcoins are sent though, clients need to be able to find out how much speed (on average) a certain fee gives you at any moment in time, so you can choose accurately how much fees you want to pay.

There is an aspect of this which is textbook tragedy of the commons, but it's kind of abstract. The scarce public good being consumed is the bargaining power of miners. As long as miners have that power, they can demand enough fees to fund the mining required to secure the network. If a miner accepts low-fee transactions, he consumes this bargaining power (since then senders will have no reason to include fees) for his own benefit, making everyone worse off (miners who are now unable to collect fees, and Bitcoin users who will now not have a secure network).

I simply don't see what the big problem is regarding this issue. First of all, Bitcoin has plenty of time to grow and mature before the lowering block rewards could be a serious problem. The most relevant indicator here is the amount of transactions in the network. If we're still at 6000 per day in 2020, we could have a problem. However in that case nobody cares, Bitcoin has not gone anywhere and is even more insignificant than it is now. So it really doesn't matter if Bitcoin is extremely secure at 6000 transactions per day or not, it matters if it's extremely secure at 600 000 transactions per day.

I claim that it will be secure at 600 000 transactions per day and it doesn't matter at all when we reach that target. At high transaction counts and low block rewards there will form a market where transaction fees are everything. Users can choose to add their transaction to the next block by paying what everyone accepts or if the user wants to go cheap and slow, then pay what at least someone accepts. Overall transaction fees will be much higher than they are now, although still much cheaper compared to banks, paypal, visa or anything like that.

I will not comment on the proof of stake thread separately because that is somewhat related to this. Proof of stake and the variations of it are interesting thought exercises and should be experimented on. Bitcoin should not be mixed into that, arguing about proof of work is pretty much the same as arguing about Bitcoin's gold-imitating monetary model. Some think they will work, some don't. Cryptocurrencies will be very competitive in the future so if you can make a better one, be my guest.

I see very low likelihood of Bitcoin having problems with 51% attack, especially now that mining is becoming more decentralized, not less. Executing a 51% attack first of all requires matching 100% of the current hashing power and beyond, if the current hashing power is decentralized enough and not very vulnerable to ddos attacks. However, the most important thing to understand is that Bitcoin hashing power is closely related to how much it's used and that will continue to be the case.

Some may think that the current Bitcoin hashing power is a lot but this is just the very beginning. We already have more hashing power than the top 500 super computers combined and we're running a network with just over 6000 transactions per day. And the hashing power seems to be growing, even now.

Imagine a network with 600 000 transactions per day. The price would be on another level and thus our hashing power as well. Bitcoin might be weak to a strong well-planned attack from certain entities on this planet right now but at 600 000 transactions per day I bet that there is no one on this planet that can do anything to Bitcoin even if he/they wanted to. It would be easier to shut down the Internet, or all electricity grids for that matter.

Bitcoin hashing power has been well ahead of the interest/motivation to attack the network. Major players don't really see Bitcoin as a threat right now and when they do, I imagine it will be too late to do anything about it. I could be wrong though and that's why I agree that Bitcoin needs competing forks. Putting all our eggs in one basked wouldn't be smart.

I'm not at all sure that there will be pricing power in the bitcoin mining market in the future.

What I am saying is that if low block reward and low tx fees make people start abandoning mining then the remaining miners will start to pick up some pricing power. This will allow them to raise prices on people who are very concerned about getting into the next block or two. The now higher tx fees and the low barrier to entry you mention then brings in more miners reducing the pricing power from running away.

The only part I disagree with is the constant returns to scale. In order to exercise their pricing power the bigger percentage miners have to decline low tx fees. Small miners get to take the large and small fees.

Brief example. Suppose one 10% miner and countless tiny miners. The 10% miner guesses/experiments to determine that if they refuse to include tx with less than .01 fee 8% of people will pay it to keep their chance of getting in the next block near 100% instead of near 90%. This costs the 10% miner all the small .0005 fees, but the make much more this way because 100 times .0005 is less than 8 times .01. But the small miners still get all the .0005 and they get the .01 fees.

The big miner does it because they would get more than otherwise and the small miners get a free lunch out of it.

I don't know exactly what technology will look like in the future, so I can't say precisely how you are confused. But I know you are confused. Current hashing technology is constant returns to scale with extremely low entry costs. Under constant returns to scale and low entry costs, any temporary build up of pricing power will lead to rapid entry and rapid dissipation of pricing power.

You need high entry costs to maintain oligopoly (i.e. market power) in a long-run equilibrium. What is the potential source of large barriers to entry? Barriers to entry would require fundamental changes in technology (for example, mining occurs only within a handful of companies using proprietary chips that are not marketed for external use). Even if this happened to be the technological outcome, a closed technological setting like this is highly likely to degenerate into monopoly.

There are financial incentives to small timers getting in on mining whenever anyone has any pricing power. That's pretty damn good. I understand that magic could probably do better.

I didn't catch this right away, but due to this poster's obnoxious "you are all idiots" attitude just want to point this out (because, hey, with that attitude, you deserve to be scrutinized)


Fee revenue is 0.03 BTC per block.
Fee submission rate is 0.001%.
Fee submission rate needs to increase 1666-fold.
OK, so far so good, but then "we need to move fees on sends from 0.01% to 16%?"
*face palm* You just said Fee per block = Submission rate * Fee rate. Why the hell would you propose increasing the fee rate instead of just increasing the submission rate? Or rather, allow it to grow on its own as Bitcoin gets adopted? Having Bitcoin use increase 2000-fold isn't much of a stretch considering how limited its use is now.

I do hope we'll see the day where people are not forced to work in order to have decent food, shelter, and iPads ... I'm really excited at the future. Still, I feel that we'll always have money, even if lots of stuff will be available to "everyone", at least in modern countries.

There is no need to be so abrasive.  If none of us had anything left to learn about Bitcoin, there would be no point in visiting this forum in the first place.

Anyhow, I couldn't find anything in Wikipedia or elsewhere that proves me wrong.

This is why I don't see total hashrate as a Tragedy of the Commons:

In a Tragedy of the Commons, each additional sheep grazing the field has a negative effect on the supply of the public good.  The long term result is total depletion of grass, ie. supply drops to zero.

In the Bitcoin network, each additional transaction has a positive or at least neutral effect on the supply of the public good (hashrate).  In the long term, hashrate might decline but it's unlikely that it will ever drop to zero.

In my opinion, a better analogy than a public grazing field is lighthouse construction.  This is the key difference:  Even if a fraction of altruistic shepherds refrain from sending their sheep to the field so that the grass can recuperate, it will still be depleted  because the remaining selfish shepherds will take as much as they can.  In the lighthouse case, a fraction of altruistic lighthouse builders will ensure a non-zero supply of lighthouses, even if everyone else leeches.

the main argument pro fees is that they are an efficient way to protect the network from spam.

I expect the number of transactions to rise by the time block reward gets dangerously low.

this chart is related:
http://blockchain.info/charts/cost-per-transaction


+1

this is not enough. we need to do better.

Suppose power is not concentrated enough for the effect I describe. Miners will fall out for lack of profitability leaving few miners each having more pricing power until there is enough pricing power to charge higher than minimum fees in at least some cases. Now more negligible miners are enticed to (re)enter taking advantage of the fact that there are now higher tx fees floating around and they can also take the smaller ones. This keeps us away from complete centralization.

There is a solution to this (that is fully compatible with the current chain!), but it would permanently lock-in the protocol, requiring 100% of miners and users to agree on and upgrade for any changes:

Ophan blocks can be absorbed into the main chain, adding their proof-of-work to the block including them, all while REQUIRING the block that absorbs the orphan to include any transactions that haven't already been confirmed on the main chain (blocks with conflicting transactions can't be absorbed).

More info:
https://bitcointalksearch.org/topic/unfreezable-blockchain-57647

If I have 75% I can refuse to build on the small companies blocks and they will never stay in the longest chain.

Pretty much. You could argue for some heuristics to detect which blocks are the attacker's and reject them, but those are unreliable and go against the spirit of the blockchain.

Of course, the network is screwed only if the monopolist is an attacker, rather than a benevolent, ruthless dictator. But the latter isn't too good either.

Which is why I think overtaking the network should be harder than obtaining 51% of the hashrate. It should require obtaining 51% of the bitcoins. (And, I recently had an idea how in a proof-of-stake system, even someone with a majority of coins can't overtake the network. I'll need to think about it).

But... with a 51% attack..... Fuuuuuuu!!!... 
Is this really how this works? 51%, reject any blocks but your own, and the whole network is screwed??

That's my point, mining should be incentivized rather than volunteer work.

I give up.

It depends on how well we can identify the monopoly's blocks and whether we choose to penalize it if it misbehaves. But in principle, as long as the monopoly is in power, it can simply reject any blocks by competitors. Therefore, nobody will try to mine. Which also means the total network hashrate (and hence the difficulty) is its own hashrate, which is as low as it pleases - as long as it's not so low a single entity can easily overthrow it.