Topic: The Bitcoin consensus mechanism is incorrectly labeled Proof of Work - page 2. (Read 4339 times)

The number of delegates won't be fixed.. Hence why I posted about the n delegate model and how it prevents top down corrupt entities from retaining control within the voting framework. N is determined dynamically in the same way. I believe min is 101 and there is a max number that isn't outrageous.

Miners or delegates or validators or whatever will expend resources only to the extent justified by transaction processing profit margin, which delegates in DPoS will also do. In fact DPoS may well have high profit margins because the number of delegates is fixed, making it a closed market.

So perhaps less electricity, but if so then more resources expended on something else (politics most likely).

(This assumes that the coin distribution phase of Bitcoin is over or insignificant, which must be done to meaningfully compare with DPoS since DPoS is incapable of distributing coins at all.)


I don't agree with your argument that your argument is conclusive. You need to show that economies of scale are net positive at the economically relevant scale, which depends greatly on many undetermined factors.

It consumes less electricity.

Since I already argued that one can't mine with lower economies-of-scale without losing hash rate share over time

You are confusing miners and users. Perhaps they could be the same, but they need not be. (Indeed satoshi's original design more explicitly divides the two than current thinking among many prominent Bitcoin developers.)

r0ach's claim was that pooling of mining in practice makes the consensus system of Bitcoin inherently delegated, but that is false because miners need not pool, and certainly need not pool with one of a fixed set or even a fixed-size set of pools. Thus this is different in structure from a system where delegation is required. (It is true if you posit that Bitcoin will certainly be 51% attacked, as I think you believe but can't prove, and turned into a centrally-controlled system instead, where pools or whatever they are they are called in that outcome enforce membership, but then why bother with the delegation argument, just show that this is certainly true and prove Bitcoin useless unconditionally.)

I specifically doubt that the outcomes will be the same in a system with permitted delegation to an open set of delegates (pools) as opposed to a system with required delegation to a fixed size set of delegates. But if r0ach wants to try to prove the same outcome in his war game model, he is welcome to do so, and I will read his proof with interest. Unfortunately if he does that, then he will show no advantage to DPoS over PoW, which undermines his original argument.

For his original argument to hold he has to show both that the distribution-of-power outcome is different between delegates in DPoS and pools in "delegatable PoW" and that the DPoS outcome is (in some specified way) preferable.

For his original argument to hold he has to show both that the distribution-of-power outcome is different between delegates in DPoS and pools in "delegatable PoW" and that the DPoS outcome is (in some specified way) preferable.

Fact is that mining will become ever more centralized in Satoshi's design because of the economies-of-scale of ASICs and electrical power. I believe there was maybe even a research paper that proved something along these lines?

The fundamental problem is the mining is done for profit. For as long as that is the case, ASIC farms (or Larry Summers' 21 Inc. economies-of-scale) and subsidized, industrial/government/utility scale electricity will rule.

Also due to bandwidth issues and that every full mining node has to validate every transaction, scaling transaction volume will force centralization.

Also your argument about sacrificing cost is nonsense, because the low cost leader will take hash rate from the others over time by reinvesting higher rates of return.


Edit: Smooth has a valid point if no entity (or collusion of entities) has 51% of the hash rate because then someone could sacrifice mining losses in return for censorship resistant way to post transactions to the block chain. So in that sense, my word "nonsense" is incorrect and I apologize. But the huge glaring flaw is that once the State can regulate 51% of the mining power (which is destined to be centralized), then Smooth's caveat no longer applies. And this is my overriding concern, so that is why I often downplay this caveat that smooth points out.

Edit#2: however if hashrate is very large then smooth's caveat is really pointless because who has enough hash rate to push their transaction onto to the block chain without a pool. And again Satoshi's design doesn't enforce that pools must allow getblocktemplate. If your hashrate is not too small, you can just mine on any pool that offers getblocktemplate and wait a long time until you win a block solution to insert your transaction, or just mine a long time solo. Many could potentially join together to pool their resources to mine at a loss to have ready access to censorship resistance, but unless you are using P2Pool (which can be attacked with share withholding attacks) then the State might target your pool server (but again I think it is easier for them to just target 51% of the hash rate for regulation requiring all transactions to carry KYC, since you might place your server behind an anonymity network although this will be very difficult to do in Satoshi's design because of the bandwidth requirements).

Except as your analogy applies to Satoshi's proof-of-work design, then the free parking is not accessible by anyone who has a car because it is on the top of a skyscraper[1]. The only people who can access this free parking must either have a helicopter or they must pool their resources to buy one.

Consider the analogy of a parking garage in a very convenient location (say right next to a popular theater) when there is free parking available a short to moderate distance away compared to the situation with the same garage but no free parking available at all. In the first case, the garage may charge only a nominal fee and nearly everyone (or conceivably everyone) might pay it for more convenient access to the theater. In the second case, the garage will charge the maximum fee possible until people stop going to the theater at all.

Ooops... I meant DPOS. 

Sorry, blame Friday.

Not even going to comment on that since I don't think you read or understand anything in this thread.

As for your naming idea, I'd agree with Delegatable PoW.

'delegated' means you must delegate, you cannot mine yourself.  That's the idea of trusted super node systems like bitshares.

The argument against my statement, is that as long as you can opt out of delegation, even only one person out of six billion choosing to solo mine with a trivial hashrate, then my statement would be false.

A common complaint of DPoS is that it can't be decentralized because it has the word "delegated" in the title, yet anyone mining with a pool in PoW is doing the exact same thing.  Satoshi made the claim of one CPU, one vote, yet you're delegating your vote to the pool owner in PoW pool mining.  Some argue you are not delegating in PoW because you can solo mine.  While this statement would be technically correct, the miniscule portion of the Bitcoin userbase able to do so makes it functionally infeasible.  It's only a question of what percent of the hash rate is being delegated at any given time.