Topic: Why the Bitcoin rules can't change (reading time ~5min) - page 6. (Read 12031 times)

Also, even though I would be surprised if I woke up tommorow to find that bitcoins had gone up to $1000 per coin, I would not be surprised if a whole lot of people's idea of how high a full-node bar could "reasonably" go suddenly jumped upward, maybe quite a bit.

So I think if the people who claim to "need" a larger block size limit to accomodate the lucrative large numbers of users they plan to bring into the system (but cannot bring in today due to the block size limit being in their way) put their "need" quantitatively into bitcoin by buying more bitcoins that could help a whole lot.

How much per bitcoin worth of people are you planning to bring in? Isn't it in your interest to hold lots of coins already yourself before bringing them in? Show how determined you are to bring them in by putting your money in the blockchain and quite possibly that will grease the wheels of upward blocksize movement quite a bit...

...And I have to admit, recent exchange rate trends do seem to indicate serious interest, and are every day making larger and larger upgrades of hardware and bandwidth look "reasonable"...

On the other hand, we hit $32 before on less transactions / smaller blocks than it is taking now to get there, so current levels don't really seem to compellingly call for more block size than we had last time we hit $32.

-MarkM-

Nope. Just me.


Peer review + github making it extremely easy to spot.


Because I'm not a peer anymore and I have no say in what the rules are anymore. Right now if the network wants to keep me as a blockchain validating, transaction propagating, block validating peer, they need my consent to download a new client. If I can't be such a peer anymore the network will give a rats ass about what I want because they'll control all the peers.


Well there's the rub then isn't. What if just a very small group of people has these means? retep seems to make a pretty good case for exactly that happening if the block size limit is increased.

I can't afford to read 100+ posts on this subject again, so I'll just address OP. Pardon me if I'm repeating something that has already been said.


Hazek,

It seems you want Bitcoin to be an absolute trust-free system to anyone, even people running lame machines which can't handle the number of transactions per second a serious payment network would generate.

But... how do you even know, for sure, that the code you're running really does implement the contract you claimed to have "signed"? Is it really trust-free? Or do you trust on the fact that, being open source, somebody with the skills to validate such code will do it, and the day some contract violation is inserted in the code, it would be quickly spotted by code-review? You see how that's not "trust-free" anymore? Very few people, even among computer specialists, are really capable of digesting Bitcoin source code. It's just a tiny "elite" if you will. And that's enough for you, for me, and for everybody else, apparently.

A 4000tps blockchain would be pretty much the same thing, with the difference that, considering how optimized Bitcoin is, it would be much easier to run a full validating node doing 4Ktps than it is to dig into Bitcoin source code and understand it. A much higher percentage of people will be capable of running full nodes than understanding the source. If you can trust the openness of the source code to provide enough reliability for you to use it, why can't you trust the openness of the blockchain to provide the same reliability? Actually, with some effort and expanses, you'd likely be able to finance a 4Ktps full node only for the purpose of validating the chain yourself. Or, at least, get together with like-minded people and finance a full node for that purpose.

Take wikipedia for instance. It's gigantic, fully supported by its supporters donations, and some researchers claim it is more reliable than old-fashion encyclopedias. Nobody can validate its entire content. For every complex article on a particular topic, only a small minority of people in the world have the knowledge to keep it correct, and from these people, only an even smaller minority will actually do it. Yet, it's openness provide this great reliability: http://news.cnet.com/2100-1038_3-5997332.html

The blockchain is open, and as such, it can be validated by anyone with the means to. That's more than enough to trust it not to change to something "evil" without major repercussion.

Nah, generators can be bought or built, even ones that use wind or water instead of trusting the oill barons; and we can cobble together an IP network out of parallel port cables and serial port cables from igloo to igloo if we have to.

Or we can just buy our electricity and bandwidth from a different provider; plus we can audit/verify the number of kilowatt-hours we are getting, the quality of the alternating current's adherence to the number of cycles per second we ordered, and check that the blockchain our internet provider is giving us verifies and that the blocks it claims nodes out there somewhere are sending us verify. So we don't need to trust either of those providers, we can verify what they provide.

-MarkM-

Ugh... 

Like the people who provide you with E-L-E-C-T-R-I-C-I-T-Y, or who run the I-N-T-E-R-N-E-T, among others?

Zero trust of course meant zero trust in other people, not absolute zero trust..  Do I really need to spell everything out?

This argument quickly reduces to bringing the goat to the market to trade it for potatoes. Zero trust is an illusion. Even now, you trust you'll have access to the means to access the internet (electricity, for example), and you trust that the internet won't blow up, rendering your bitcoins worthless.

It's about managing trust, not about reducing it to zero.

Maybe the incentive is "multiply the fee your local (or trusted, or least-hated, or least-distrusted) blockchain-audit corporation charges you for consultations regarding the verity of the blackchain by the number of such consultations you expect to or want to make, and compare that cost to the cost of running a node so that you can audit it yourself"

-MarkM-

Precisely. Right now when I advertise Bitcoin I advertise it as zero trust. I'm afraid, if the little guy can't validate what rules are validated anymore, I wont be able to do so anymore and you can bet many many users will not like this at all to the point that they'll leave, just like I will.

We already have trust the big guys systems and look how that's working out. And if anyone is going to try and discredit this position as some sort of ideological grandstanding I invite you to look up the genesis block and read what Satoshi wrote in there and learn what exactly his motivation was when he started Bitcoin.

This!

But for this to really kick off, there should be an incentive, like the miners have an incentive. Any ideas? And we can finally close the book on the last hypothetical problem with BTC and go worldwide with this baby.

Which boils down to is it or is it not a zero-trust system?

I suppose you'd argue it is a limited entry level zero trust system, a system that is zero trust for the megacorps and any sufficiently wealhy nations or other sufficiently well heeled players, but for normal folks is just another trust the big boys / trust the grownups / trust Big Brother system?

-MarkM-

This sovereignty you mean is mostly an illusion. The real power you have is that if the rules change in a way you don't like, you can sell your coins and go use another system that works the way you want. this is not substantially different from just sticking with the old version. The specifics do differ though. Whether you're actually validating things yourself only affects whether you have to trust what others say about the system or if you can verify it for yourself.

Mike, like I said a hundred times already.. I don't care how it's solved and am willing to accept almost any rule change as long as I keep my Bitcoin sovereignty. That's all I care about.


BTW Bitcoin sovereignty doesn't mean I just want to be able to validate what rules are used when transactions are validated and blocks created and added to the blockchain. It means that these rules also can't change for me without me downloaded a new set. Right now it's extremely hard to do a hard fork if you don't have near complete consensus because of the economic ramifications. You need almost all users to agree. But once you take away full clients from merchants and regular users, rule changes can happen at any time because you don't need them to download anything anymore because they aren't actually validating anything anymore. Yes you can probably still spot a rule change, but you can't affect it by being part of the network and refusing to download a new client threatening a hard fork.


But hey, as I said, I don't care how this is achieved. If a light client can give me the same sovereignty, I don't care, as long as it does. As long as I have to give my explicit consent to a rule change, I'm happy.

These discussions are getting silly.

hazek, even if running a full node gets too expensive for you (which I don't think is likely), you don't have to pay for a node entirely by yourself. You can team up with other like minded people and split the costs. Using assurance contracts and a computer in a remote datacenter capable of trusted computing, you can even do it with people you don't know and don't trust. By the time this becomes an issue, the software to make all that easy will probably have existed for some time. If not then again, you can team up with other people and fund it. I'm sure many people would feel the same way as you and be willing to contribute.

But even at high transaction rates, the cost of a full node is not likely to exceed the spare funds of an employed person with a decent wage anytime soon. Not unless you believe Bitcoin will become more widely used than Visa in the next couple of years, which would be absurdly optimistic even by the standards of our community

People on 56k modems can of course use SPV clients even at high transaction rates. You know exactly what rules they are enforcing because, assuming you are patient enough to download web pages over your dialup connection, the ruleset being enforced by the miner majority will be documented somewhere! If you disagree with those rules, well, that sucks to be you because then you're no longer on the best chain even if you run a full node and are essentially using a different currency.

In short, even if Bitcoin becomes enormous, I can't see a time when hazek is unable to check what the rules are, even in the unlikely even that he is forced to rely on SPV mode.

Yup. A non-trivial (more than a few percent on each side) mutually exclusive fork is the worst possible outcome for bitcoin. From one currency is two, and all older coins can be spent twice.  There are enormous incentives to achieve universal consensus one way or another.

I'd like to get back to the original post: could you please elaborate on what happens to a regular user after the fork? What if I stick with the previous version of full node client and others switch to the new one? Will I be able to transfer BTC to these users? Or does the fork mean that the current money supply gets divided between those two versions?

No. Banks are the goliaths they are solely because they have the politicians in their back pockets. A Bitcoin mining lobby is a hard thing to imagine. This makes new entry prohibitive, not merely expensive. Watch "Bank of Dave."

Nope, just the ethical central planner libertarians who think their interpretation of what is voluntary and what is coercive is the objectively best one. Rothbardians and Randians always encounter problems because of their moralizing, but Misesians and especially Hayekians can understand what Bitcoin is about. Bitcoin isn't about monetary sovereignty; it's about not having centralized power over money, allowing the free market to work. The idea of "objective" sovereignty is a minarchist holdover stuck in the old statist paradigm. In a stateless system the benefits are myriad, but being able to "control your own money" in the sense of controlling the system itself to any significant degree is not one of them. The endpoint of such reasoning is going off to live in the woods as a hermit. That is not what libertarianism is about.

+1 You are absolutely correct.

I have a lot of respect for Hazek, but if I read the OP correctly: that Bitcoin can't change unless 100% of its (full node) users agree, then it will lead to paralysis. In the worst case, because 100% agreement can never be achieved, Bitcoin will split into different coins, or fail completely. There needs to be some way forward even if a *small* percentage of users object. The real world is not perfect, in case some people are forgetting...