There are a few prerequisites needed before people can start producing economic surplus.

Tribal people in rain forests haven't yet figured out how to not routinely murder each other, so learning how to not respect property is still a ways off.

The value of money comes from future economic output.

The majority of entities that will produce the most economic value in the future is the majority that matters.

In many cases, people who have accumulated a large amount of money in the present have done so because they have a high capacity to produce economic value and so in their case they contribute significantly to the economic majority.

This is not true in all cases, however. Someone who obtained large amounts of money through luck or via processes which are not repeatable in the future don't contribute as much to the economic majority as their current holdings would suggest.

This is the primary flaw behind who try to frame debates in terms of "rich" vs "poor". They aren't being sufficiently precise.

Who decides what parameters to use for the exponential cost ratio?

Miners dont really control the  protocol like that.

Broken clients would recognize their fake blocks, but only until we fix the broken clients:

https://gist.github.com/justusranvier/451616fa4697b5f25f60

There are people in the world who perform elaborate dances every week because they are absolutely convinced that without that ritual the sun won't come up in the morning, or winter will never turn into spring, etc.

If people want police, roads, and other stuff, then somebody will provide that service.

The idea that only the state can provide those things is a fairy tale. It's kind of embarrassing to live in a world where so many adults still believe in those kind of myths.

Cool story.

The only reason we have to talk about spam is because the resource allocation of network bandwidth and storage isn't handled very well.

Nobody is ever going to agree on what is or is not spam, so a more productive solution is to make whatever changes to the network are needed to ensure that everybody pays for what they use.

Once that condition is achieved, it doesn't matter how many resources people use.

I agree that if the least secure and reliable way, out of all the other possible ways, of storing the blockchain was chosen, it probably wouldn't work very well.

Did you have anything else to add to the discussion?

Can you rephrase your comment in terms of how the number of nodes is related to the number of copies of the blockchain in the world, and how both of those affect the ability of an attacker to perform double spending and censorship attacks?

I care because there's an appalling lack of rigour going around and it's not helpful.


I think that's a good start. The question can then be translated to "what are the minimum requirements needed to prevent double spending and censorship?"

I suspect the answer in terms of the number of world-wide copies of the blockchain needed has a lower bound of 1, depending on our ability to create efficient fraud proofs and to implement better privacy behaviour in wallets.

Is he right or wrong about that?

Via what methodology would we test that hypothesis before arriving at a conclusion about its validity?

I wish that Pierre Rochard would stop pretending to be a proponent of Austrian school of economics and would openly come out as a Keynsian/interventionist. It would be a lot more honest.

I don't believe any who claims or implies that they do not fully understand this point.

Everybody knows what the real problem is, they just pretend that it's about the signatures instead about the spam posts because that's the only way to pretend they are doing is vaguely defensible.

Let's observe the massive human tragedy caused by giving people free money, and propose a solution that involves giving them more free money.

Somebody is buying a lot of bitcoins today.

This is a self-correcting problem.

The #bitcoin-wizards have fallen into that trap before, when the btcd team announced the results of their benchmarking on large simnet blocks.

Their complete system measurements were quite a bit slower than what the individual component microbenchmarks suggested.

If you look at the comments made regarding their blog post, you'll see the same dynamic playing out as is here.

That's one way to do it, however even this can be shortened.

Right now with all the blocks < 1 MB it's not really a big deal to supply the entire block to prove that the referenced transaction doesn't exist, but it'd be nice to not require the entire block especially for when blocks are larger.

By adding a rule to new blocks that require all the transactions to be ordered by their hash, you don't need to supply the entire block to prove that the transaction doesn't exist.

It would be good to have that ordering requirement in place before blocks are allowed to grow to make sure that fraud proof size is bounded.

Thoughts on how fraud proofs could make it possible for SPV clients to reject an invalid chain, even if the invalid chain contains the most PoW:

https://gist.github.com/justusranvier/451616fa4697b5f25f60

(some modifications to the Bitcoin protocol required)

Maybe it's a good time to start talking about enabling fraud proofs that would allow SPV clients to reject invalid blocks.