Topic: Bitcoin 20MB Fork - page 62. (Read 154787 times)

No, that is not the way to do it at the conceptual design level.  Doing so is the best way to ensure a completely counter-productive waste of time.  Once a realistic end-goal is arrived upon then comes the phase of breaking a problem into it's more simple components and working on them.  No wonder I cannot see eye-to-eye with almost anyone here!

One challenge/complexity would be that there is an time-bound asymmetry between the network cost, and the pay for sending (or for relaying) a transaction.

The Bitcoin network cost = Data Size * decentralization.

The transaction must be stored on every node and forever, whilst the fee is paid once and to only one recipient.  Bitcoin solves this using a free market solution of the block solving lottery.  The assumption here is that the block reward is sufficient for the miners to "tip out" to the nodes.  To make sure that they keep at it.

Peter Todd twitter-quipped that the block size is a barrier to entry, he is right, but this is balanced by another barrier, the data size.  This grows at the rate of the block size.  So the bigger the block sizes, the more transactions can be processed, but also the greater ongoing cost for storing all those transactions for an indefinite future.  So a cost-payment asymmetry.   The one-time payment must cover costs for an indefinite future.

Getting this balanced is not easy within the paradigm of a protocol.  This is how we became stuck with a block size limit which we hope to outgrow over the next few years.


Yes, that is the way to do it.  The complexities get simplified at the time the code is written.  Unless they can't be simplified, and then the issue doesn't get solved.

Uh.. The Bitcoin Foundation is doing no such thing. I think you are mistaken.

Baffles me mate.

why do you only want to make that change if it is too late?

when would you do the change? you know that a hardfork needs time and i strongly believe that the media would love stories like "bitcoin collapsed because they have 10k users now"

Sure, but please first demonstrate how there's currently not enough of it.

The article looks like warmed over Mike Hearn vomit from back when he was talking 'assurity bonds' or some such nonsense to justify getting rid of the block size limitation.  He just didn't couch his argument in wall-to-wall libertarian hucksterism.  To bad for him else it might have worked even back in 4 years ago when the 'crisis' existed much as it does today.

Justus's brilliance is that he sheds complexity by simply ignoring it.  Much like solving an algebraic expression by simply striking off any terms which contain radicals or other complexity.  And it works like a charm here on trolltalk.

The trouble is that we don't have 'free markets' and have no path to get there.  We live in the real world, and in what passes for 'free markets' that we have Bitcoin is pretty certain to arrive at an end-state represented by pretty much any other similar-ish solution.  PayPal, e-mail, etc.  The global internet's hold on it's claim as a free medium of information exchange has undergone significant regressions and is tenuous at best and certainly not solid enough to base a meaningful disruptive monetary system on.


Ha!  This is as funny as a crutch!  After years of doing everything possible to get rid of the pesky P2P nature of Bitcoin, TBF is just realizing why it is important and is doing a minor about-face.  Just long enough to get their exponential growth fork in though.

In that case let's lower the block size to say 50KB, get some more competition.

Probably the guy is mainly targeting me as the most vocal of the sidechains proponents.

The concept of a sidechain is extremely simple.  It is just a subordinate chain.  It derives it's value be absorbing and locking BTC from the main chain.  It impacts the main chain in exactly as would a typical user (albeit relatively large and active one.)  To the extent that a healthy sidechains ecosystem would greatly increase demand for Bitcoin without bloating it, and would be able to afford large transaction fees since the fees would be an aggregate of thousands of economic transactions rolled into one, sidechains would be of huge benefit to Bitcoin.

The problems this would solve would be to really make the power and independence of Bitcoin available to a lot more people without compromising the defensible nature of Bitcoin.  To the MP crowd this may not be of significant value, but it is to me.  Beyond that, what makes Bitcoin valuable is it's ability to be 'disruptive', and the more people who can use it, the more disruptive it is.  If MP and crew are not seeing this then greed is obscuring their analytical abilities as far as I'm concerned.  True, the more disruptive, the greater the threat of attack, but we already need to be planning for the maximum level of attack anyway.  Bitcoin's days of security through obscurity are very much on the near horizon.

Part and parcel to the 'accessibility' issue is that some uses of a crypto-currency are mutually exclusive to one another.  A good example would be whether they play ball with the legal authorities.  For many uses there is not much reason not to 'register' and be monitored by the powers that be.  Some sidechains may do so.  Other sidechains may not.  Absent this flexibility Bitcoin, as a monolithic one-size-fits-all solution, will have to make a difficult choice if forced into a corner.  My expectation from watching the ecosystem over the years is that this will be a very easy choice for most:  ask 'how high' when instructed to jump.

Another thing about sidechains is that they would introduce a large degree of extra complexity in trying to track individual user activities.  This even if there was no effort put into fostering this protection.  If there were, the problem could be made fairly insurmountable for attackers.

Bitcoin already has enough baggage and design deficiencies to where a better 'reserve currency' could be created, but still here in early 2015 is a workable solution with a significant first-mover advantage.  That potential pretty much disappears when TBF finally gets their exponential growth fork in.  I think there is a pretty good chance that some people see full well the power of sidechains to work, and work fairly well, and are in something of a panic over it and that this may be associated with the latest push to get the exponential growth fork to happen ASAP regardless of the necessity on technical grounds.

Welcome to rational economics (not the usual bullshit you read in the mainstream press).

Justus did an awesome job.  Sadly the trolls (davout, danielpbarron etc.) have nothing to contribute besides noise.  Here's hoping they play their spend games when the fork happens and lose expensively, it'll be a valuable lesson they actually paid for.  However, it's pretty clear to me that they will pussy out.

I don't know what a "side-chain" is, and have certainly not been advocating it as a solution to anything. To imply that it would "solve" something would to be imply there is even a problem, which there isn't. The onus is on proponents of the hard-fork to demonstrate a problem, and so far they have not. So far the argument is "waaaah I can't be a freeloader anymore because the transaction space is limited and I have to outbid the competition," and I'm just not convinced this is a problem.

As someone who runs a full node, I am very familiar with how difficult the process is. It's hard enough to sync with the network as it is; 20 MB blocks would be impossible. It would take months if not years. We don't all have access to whatever military grade USG hardware Gavin does.

Use of the blockchain is like use of automobiles; all the poor idiots have enjoyed use of these things despite their being utterly unqualified to have anything but a chain around their neck and a whip on their back. Enjoy it while it lasts, because when the serene republic takes over where FIAT has failed, you will have neither free access to the blockchain, or even a car to drive. You will be lucky if we will even have you as a slave, because from what I've seen on this forum, you probably couldn't even do that properly.

If there was an allocation to node running as proposed, it would help.  Since nodes are needed for mining, it was envisioned that the mining rewards would do this.  

In economic solutions to avoid centralization, look for the Nash equilibrium and to avoid solutions with advantages weighted toward incumbency.  Justusranvier has a pretty well reasoned description of what that might look like.

Creating a market for node running would not be a bad thing, if done with incentives favoring distribution in the design.  TBF has a sort of small incentive program going:
https://getaddr.bitnodes.io/nodes/incentive/

It is an off-network incentive, so maybe we'll see what happens with it..

So they have been lying this whole time? 

Don't know why people are giving him a hard time
I honestly think the fork from 1 megabyte to 20 megabyte is a great idea.
I have even created an altcoin with max block size 20 megabytes https://bitcointalksearch.org/topic/m.10054857

I missed that one. I'll need a lot more coffee to sort through that api proposal. That sounds like the right direction we need to be going.

Fair enough. I perhaps did not give the author the credit that was due. He is quite aware of the problem. I need some time to read and digest this https://bitcoinj.github.io/working-with-micropayments and also some time to figure out if, even if we had unlimited free micro payments, there is a means of dealing with hold out problems.

*edit* Wow this is huge if he is right. If he is and the hold out problem can be sufficiently mitigated than we dont just have a model for solving the max block size problem, this will by default also solve the problem of how to decentralize the entire internet.

*edit2* wow wait thats you. you wrote it. lets talk. pm me. better yet can we skype some time?

It would be great if you could read the actual proposal and show where the problem is specifically, rather than just list your a priori assumptions.

I'm all for "letting the market decide" unfortunately we have no means for pricing the cost of transactions. Sure the transaction author can negotiate with a miner but he has no way of negotiating with every node operator in the world and every single one of them bears a cost for his transaction. Even if a node opperator has sizable bitcoin holdings it is unlikely that the befit of the increased value of his bitcoin as a result of his running a full node will outweigh the cost, so this cost is effectively "socialized" and he is expected to bear it for the sake of altruism.

The point is, we all would love to avoid central planing, but without the ability to make a market we have no alternative. This article presents a false choice, that of central planning vs allowing the market to decide, but there is no market here. The best we can hope for in terms of creating a market to address this problem is competition among crypto currency central planners, then the market can decide which central planners are producing the best monetary policy.

Dont think that im some sort of socialist. I'm as pro market as they come. This is just the unfortunate reality of the situation. I wish it weren't so.

isn't the new bitcoin core all about the headers first/parallel downloads thing - getting the blockchain in about the same time as the torrent (3 or some hours ) now ?

also, reading Justus Ranvier's  'Economic Fallacies and the Block Size Limit, part 1 and 2' makes a compelling case for thinking about all this stuff differently to the gavin approach -

https://bitcoinism.liberty.me/2015/01/21/economic-fallacies-and-the-block-size-limit-part-1-scarcity/

https://bitcoinism.liberty.me/2015/02/09/economic-fallacies-and-the-block-size-limit-part-2-price-discovery/

Maybe it's steganography.