Topic: Reversion to the mean - page 2. (Read 2312 times)

We could've saved a lot of time if you had told me up front you're with the gang that proposes nodes will eventually be run in datacenters.

https://en.wikipedia.org/wiki/Principle_of_good_enough

You are presenting a false dichotomy. If running a full node costs several thousand dollars and 1GB/s bandwidth the Bitcoin is worth a whole lot to a whole lot of other people. It being worthless to you is your call, not an absolute.

You don't need to run a full node, you might want to. I want bitcoin to totally replace cash. We both have our dreams

I'd agree. I don't need brg444 to run a node. I can run my own. If brg444 needs brg444 to run a node, then brg444 needs to pony up and get it run.


Miners the group is made up of many individual miners. An individual miner is on balance likely to do what is best for itself. It is not naive to rely on an individual acting in their own best interests.

Miners that mine really big blocks to try and force out smaller players face the risk of a smaller player mining a smaller block faster and orphaning the big block. Thats a really simplified example. Reality is far more complicated. So much so that its very difficult to know for sure exactly what will happen - that to me is the leap of faith. Before Peter R went bonkers with his GIF he did put together a paper which, albeit not perfect, was a good model. Yes, spherical cow etc thing is its the closes to a model we have seen, and it intuitively makes sense because it is analogous to well understood economic theory on supply and demand.

What are the chances that the economics around blockchain resources broadly confirm to the simple principals of supply and demand and price discovery?

What are the chances that some entirely novel and as yet previously unseen economic model independently arises that entirely screws up the whole of bitcoin?

I dunno based on balance of probability though I'd say A: That the market can figure all this shit out and that we don't need to try and impose artificial constraints.

There is such a thing. Either you are decentralized or you are not, there is no in between. Making it difficult to run a node =! centralized. Running nodes being out of reach from YOU =! centralized.

If they can't level up, they are irrelevant.

Small nodes in a peer-to-peer network aren't irrelevant in any way. They help on validating blocks and tx, so they play a major role in the whole network.

You are so dumb.

There is no such thing as "good enough decentralization".

If I can't run a node because it costs several thousand dollars and 1GB/s bandwidth then Bitcoin is effectively worthless to me.

The only way to use Bitcoin in a truly trustless and private manner is by running a node.

Bitcoin is about monetary sovereignty, not depending on corporations, banking parasites and everyone else that can afford datacenters to maintain the network.

A PEER-TO-PEER NETWORK.

What I am describing is levelling up the game. Making it impossible more difficult for you and me to run a node might result in a loss of a couple of irrelevant nodes but it does not mean bitcoin will become centralized. Good enough decentralization is good enough and the market will keep it that way. If the node count drops down to a dangerous point, the market will react accordingly to resolve this situation.

This is not a "leap of faith", this is downright naive.

I'm sorry but what you posit is straight up nonsense. The miners are not a group, there is no such thing as "their own best economic interest". It is a competitive environment and they seek profit first and foremost. Miners with the most resources will set a limit beyond what the smallest players are capable of handling to eventually drive them out of the market.

You are right that they have talked amongst themselves before, it was to organize this SPV mining scheme that succeeded in causing a fork of the network. Are you really suggesting we trust these guys to do what's best "for the greater good"!?

Precisely.

You do know what you are describing is centralization and these bigger blocks will eventually make it impossible for me or you to run full nodes.

You're right! But I would posit that they don't need to agree, they all need to act in their own best economic interest and that the free market will decide the most efficient conditions. (one might argue that they would talk amongst themselves and come to some consensus as this would be in there best economic interest. they have before, though of course they may not again).

I accept this is somewhat of a leap of faith, but the opposite requires on being smarter than the average bear. Humans are not, we are married to the idea that we are smart, and blind to all our flaws as a result.

Of course miners have different resources and letting them compete for these resources is only good for the network to improve.

Why would miners need to agree on the soft limit? They don't have to. Those with the best resources will produce the bigger blocks. That's all.

 

That was a rhetorical question.

The point is that every miners pick their limit and obviously each of them have different resources therefore it follows that they will never agree on some magical soft limit every one of them will enforce.

That's just straight up delusional.

https://en.bitcoin.it/wiki/Scalability_FAQ#What_are_the_block_size_soft_limits.3F

Should miners also decide the block time interval? Or the total supply?

After all who knows the best option right? Might as well leave "headroom".

I'm curious, how do you propose the block size could be kept down "if it turns out to be a problem". Is it up to the miners or not? What are "soft limits"?  How do you enforce them?

I know you probably have read all of these arguments but are not concerned with nodes having to keep up with the miners incentives to optimize for profits?

Do you not agree with this? Do you think big game hunters should decide on the acceptable limits of endangered species they can hunt?

You are dumbfounded about something? That needs to go on record. Usually you have all the answers

Yes you can't predict future events and so you should not take a course of action that is based on assuming something will or will not happen. Instead you should take a course of action that allows for (and positively embraces) the widest possible range of scenarios.

You have read Taleb haven't you, you know how antifragility works? You can see that imposing an artificial limit creates a fragile system? You are locked in to a particular view/model. If that model is not right then there is the potential for great loss.

Lets say we hard fork to an implementation with no block size limit. You are now in a position where you can go either way. The headroom is available if you need it, soft limits can be used to keep block size down if it turns out to be a problem.

And *everything* in between is available and possible. You aren't relying on a few good man guessing rightly, you are letting the market figure it out.

If you don't know what the best option is, and you don't however much you think you might, then leave your options open.

I don't want 1MB blocks, 8MB blocks or 8GB blocks , I want miners to have the option to produce blocks as big as is necessary at any given instance.

The 1MB block size limit was about a DDoS vulnerability. Not about the bitcoin economy. A whole bunch of people are trying to make it about the economy, and yet you keep accusing me of rewriting history.

I am a bit dumbfounded as to what your argument is... It seems you are proposing we cannot predict future events but that on the other hand we should act now to prepare for them.

From where I stand it appears you are making an attempt at interpreting what the "black swan" will be?

"Peter R: Bigger Blocks = Higher Prices"

This dubious correlation is the crux of the issue (as much as "big blockers" might deny it). The mooners want moon today, and the block size limit should have been lifted yesterday to enable that. Unfortunately, merely increasing capacity (and simultaneously opening Pandora's box for attack vectors that exploit block propagation delays) does not create organic adoption.

Why can't people wait until adoption and overcapacity actually warrant increasing block size? In the meantime, we need to address the lack of disincentives for spam/dust transactions that needlessly increase throughput requirements on the system.

I agree, we need a lightning network. So that when it becomes uneconomical to transact on chain, there is a backup. What I don't agree with is that uneconomical should be forced. Let the free market decide. Give the miners the headroom to discover the most cost effective block size. The stepped approach in BIP101 goes someway towards it, but really the system should be self regulating. No cap.

Propogation, orphan rates etc are engineering issues that will be solved when its worth solving them. Free market.

Lightning network will be implemented when its worth doing so. Free market.

Capitalism works, Unbridled it fails. Bitcoin is the reigns.