Topic: Please do not change MAX_BLOCK_SIZE - page 17. (Read 13023 times)

Larger blocks lead to it being increasingly difficult to mine anonymously, thus making it easier to authorities to control those mining.

well it's sort of rude to talk about what he means while he's reading it, so can answer himself.
but I started - I know, sorry

He doesn't mean that and in any case, a parallel chain could use merged-mining.

I think he means we should stay together and join our power to protect the net

If a >50% attack is a problem, it won't matter what the block size is, unless you're suggesting a mechanism by which larger blocks leads to authorities getting hold of hashing power..?

Sounds like a word of a wise man

Wishful thinking.

If censorship was a problem, that would be because hashing power was under the control of authorities, who would simply use that hashing power to 51% attack the censorship resistant Bitcoin. Changing the PoW algorithm doesn't help either, as new coins haven't been able to survive 51% attacks by people having fun, let alone focused attacks, for a long time.

We've got one shot at getting a decentralized PoW-based consensus system implemented. Don't screw it up.

Bitcoin can always be forked into a block size limited version if we find censorship is becoming a problem with larger blocks. Artificially limiting the block size now is creating a problem to solve a non-problem.

Eventually it will dawn on the business types that only a handful of entities are capable of consuming the whole pie, and that if Bitcoin proper can remain a rock solid core there will be many many slices of pie which will be plenty big enough to keep them busy.  At least those who's primary interest is economic.

Admittedly it is bit presumptuous for me to play business guy however.  It is probably true that one can be severely dis-advantaged by jurisdictional differences in a marketplace which is truly global.

It's actually quite interesting topic to consider.
Splitting the coinbase into forks, where one stays at the limit, while the other one goes unlimited in size, or there are more branches, with different limits...

I think technically that could be done quite easily. Almost naturally - you just let a fork go and see how the market acts upon it.
BTC would split into BTC1 and BTC2 and from that moment each of them would have a different exchange rate. At the beginning, probably both below the last common price, but later - obviously either of them is still better than any other concurrency..
Maybe that could actually work

I'll just post this again from retep's ridiculous video thread, since I don't see enough people arguing the point about Moore's law.  Moore's law means that even with no blocksize limit, we will be able to store the blockchain on PCs forever.

The blockchain is not currently increasing in size linearly in a straight line, as you can see here:
http://blockchain.info/charts/blocks-size?timespan=all&showDataPoints=false&daysAverageString=1&show_header=true&scale=0&address=

It is increasing exponentially because we are in the adoption stage where more users are added.  Once bitcoin reaches its potential market size, or once it hits a blocksize limit, the blockchain size will increase linearly.  We will never see sustained exponential growth in the number of monetary transactions that humans need to conduct.

Bitcoin merchant adoption and transaction fees have grown 1000% over last year, which is an extremely high rate that temporarily outpaces hard drive capacity growth, but this cannot be sustained very long.  Hard drive space and internet speed show no signs of slowing their exponential growth rates, so they will always outpace the bitcoin blockchain in the long run.  Exponential growth will always win out over linear growth.  People will be able to store the blockchain on their personal computers forever.  Removing the blocksize limit doesn't change that.

There was a recent article in Bitcoin Magazine that analysed the blockchain size, but most of their articles are not available online.  The article looked at some worst case scenarios, and even using really conservative estimates, 20 years from now people will easily be able to store the whole blockchain on their phone and download the whole thing in a few hours.

Hard drive capacity over time:

World internet bandwith over time:


World internet bandwidth is smashing the pace of Moore's law.  Tor bandwidth also follows Moore's law.

1980:  300 baud
1990: 14.4k
2000: 384k dsl
2010: 20mb cable
2013: 1000mb fiber

Storage capacity is also likely to have radical breakthroughs that smash the pace of Moore's law.

What I find most frustrating about these debates is that people who profess concern about centralization in the form of resources needed to run a node propose as a solution centralization in the form of restricted transaction rates and blockchain gatekeepers and nobody seems to care or notice the discrepancy.

Etotheipi proposed, and maaku is implementing, a solution to allow people who don't have the resources needed to store a huge blockchain to still be first class network citizens. If people really are concerned about resource usage why aren't they involved in this project? If there are problems in the proposal, are they offering possible solutions or at least actively assisting with the search?

When someone puts forward something as a problem, and then shows absolutely no interest in possible solutions to the problem, then it's reasonable to assume that they don't actually want to solve the problem and it's being brought up as an excuse to further some unstated agenda.

OK, so now I see some bitcoin war having a chance to pop up in a near future. People vs corporations?
It's not good, and I surely hope to stand on the right side.
Though, I don't think it's likely for me to move to the enemy camp, unless they just force me. But I will surely try to fight, FWIW, because I want to have a bitcoin node in my home 10 years from now. I mean, without a need to have also an A/C to cool it

You have brought up the bigger issue though of what people think decentralized actually is, or how important it is. For instance as jdillon pointed out recently in a pull request the Bitcoin Foundation doesn't include the promotional and protection of decentralization in its bylaws. I'm not sure jdillon had the right approach in that pull-request, but the responses of various foundation board members are interesting, especially with regards to anonymity and privacy. There certainly is a segment of the Bitcoin community that doesn't see decentralization as particularly important, and would happily sacrifice it in favor of extremely low cost payments of any size. (remember how the conference had the tagline "Bitcoin - The future of payments")

I agree with this. It would be nice if this sentiment would come out more often so that nobody would think bring up what Satoshi's original intentions may or may not have been in these conversations is a valid point.

Exactly.
So my idea is: leave the limit as is, and let the fee market to work. This will naturally create incentives for payment processors to appear and compete with each other. Just like mining pools do ATM. It even makes a lot of sense that the mining pools would become payment processors.

I've said many times before that Satoshi was a man, not a god.

Bitcoin, as originally released, had a 32MiB blocksize limit and it was implemented in a way that makes me wonder if Satoshi didn't realize what an issue the blocksize limit would be. He later changed that limit to 1MB, and did so relatively secretly.

In any case what matters is how we can make Bitcoin scale now while remaining decentralized and censorship resistant, not what Satoshi thought Bitcoin should be four years ago.

Obviously you care, or at least you think the people you are trying to convince care, because you relied on as the basis of your appeal to authority.


So it's important when it supports your point, but then we shouldn't worry about it if it turns out to be false?

For the record, Tor was not really on my mind for this.  I've never convinced myself that Tor is a sufficiently reliable framework in 'war time' environments.

Paracoin is mostly envisioned as a platform upon which to reasonably experiment with constructs which would push core infrastructure to the network's edge...and beyond if need be.  These might include rewards proportional to desired diversity in terms of geo and political situation, transaction messaging protocol, peer count, and such.

My latest thoughts have to do with 'two-pass' mining where the 'tip-blocks' are secured with suite of diverse and non-predictable algorithms while the trailing blocks are locked with high difficulty sha256.  Hopefully this could reward both those running FPGA (hopefully at home) while anyone with an investment in current ASIC and what-not could continue to participate and secure the solution.

But anyway, I would strongly prefer that Bitcoin proper remain the 'gold standard' as a source of truth in distributed crypto-curency land.  'paracoin' and things like it are only a fall-back should Bitcoin go the way of PayPal due to excessive commercialization and associated capture.  It probably understates things to sat '>90%'.  It's probably closer to '>99%'.

  Edit:  More clear about 'edge'.

It's not a lie, man.
1 MB block limit - it's there and has been there for years, whether you like it or not.
Was it Satoshi himself who added it - who cares? Who is Satoshi, anyway?
The only thing that matters is that some people don't want it to grow bigger, and they have an actual valid reasons. Very important reasons.
And you accusing me of "making up facts" - it's just silly