New video: Why the blocksize limit keeps Bitcoin free and decentralized

[d'aniel]

Personally, I'm fine with a situation where miners anonymously connect to pools that migrate to friendly jurisdictions, but that's just me.  And that's only the worst case scenario if the assumption that Tor can't scale is correct, which I've seen no argument for.  Perhaps retep or jdillon could explain this assumption here?

You know, if you are happy with just assuming pools and nodes in general migrate to friendly jurisdictions you do have a lot of options. But Liberty Reserve just showed how seemingly friendly jurisdictions don't always stay friendly.

I certainly don't assume they would.  But mining pools are a dime a dozen, and it's not really a big deal if a jurisdiction turns unfriendly and takes one or two out, as miners can seamlessly migrate to another pool in a friendly jurisdiction.  Engineering for the scenario that all jurisdictions across the planet have become unfriendly toward Bitcoin is kind of silly IMHO.

[d'aniel]

I can't see why it matters if an evil miner fills his blocks with transactions that he hasn't relayed.  It doesn't put the Tor miner at any relative disadvantage, as he's still safe receiving the block data without Tor, correct?  Am I missing something here, retep?

That's only true if Tor miners are the majority. If they are the minority, and >50% of hashing power has fast connections, doing so puts that Tor miner at a big disadvantage, and is good for the miners with the fast connection by reducing competition. Keep in mind that it's likely that large well-connected miners will offer SPV node connection services in return for you not giving the competition a chance at mining those transactions and collecting the fees: https://bitcointalk.org/index.php?topic=197169.0

I'm not following why Tor is an issue at all in this proposed attack, since they're receiving all incoming data without Tor, at the same speed as non-Tor miners.  I'm really sorry if I'm just being thick here.

[Peter Todd]

I can't see why it matters if an evil miner fills his blocks with transactions that he hasn't relayed.  It doesn't put the Tor miner at any relative disadvantage, as he's still safe receiving the block data without Tor, correct?  Am I missing something here, retep?

That's only true if Tor miners are the majority. If they are the minority, and >50% of hashing power has fast connections, doing so puts that Tor miner at a big disadvantage, and is good for the miners with the fast connection by reducing competition. Keep in mind that it's likely that large well-connected miners will offer SPV node connection services in return for you not giving the competition a chance at mining those transactions and collecting the fees: https://bitcointalk.org/index.php?topic=197169.0

I'm not following why Tor is an issue at all in this proposed attack, since they're receiving all incoming data without Tor, at the same speed as non-Tor miners.  I'm really sorry if I'm just being thick here.

The miner on bandwidth limited anonymous connections will always receive an incoming block slower than someone located in a non-anonymous datacenter.

[jubalix]

Someday people are going to look at 1 MB like they do a 1 KB. Keeping it at that arbitrary size doesn't help anybody. I'm sure there is a decent algorithm that will allow for an increasing sized block without it ruining bitcoin's decentralization.

Furthermore, I still don't understand why full nodes don't receive some compensation? Why only miners?

this is what I thought as well
...

need is the mother of invention

[d'aniel]

I'm sure there is a decent algorithm that will allow for an increasing sized block without it ruining bitcoin's decentralization.

Actually it seems pretty clear already how this can be done.  Some of us have been thinking ahead  Here's a good summary of it by gmaxwell: https://bitcointalk.org/index.php?topic=137933.msg1596626#msg1596626  Follow the thread/links back for more details.

One further improvement on this proposal is for nodes to partially verify blocks in order to make the block auditing more decentralized.

[d'aniel]

I can't see why it matters if an evil miner fills his blocks with transactions that he hasn't relayed.  It doesn't put the Tor miner at any relative disadvantage, as he's still safe receiving the block data without Tor, correct?  Am I missing something here, retep?

That's only true if Tor miners are the majority. If they are the minority, and >50% of hashing power has fast connections, doing so puts that Tor miner at a big disadvantage, and is good for the miners with the fast connection by reducing competition. Keep in mind that it's likely that large well-connected miners will offer SPV node connection services in return for you not giving the competition a chance at mining those transactions and collecting the fees: https://bitcointalk.org/index.php?topic=197169.0

I'm not following why Tor is an issue at all in this proposed attack, since they're receiving all incoming data without Tor, at the same speed as non-Tor miners.  I'm really sorry if I'm just being thick here.

The miner on bandwidth limited anonymous connections will always receive an incoming block slower than someone located in a non-anonymous datacenter.

Um...  These attack blocks would be received on non-anonymous connections, as this behavior is indistinguishable from a non-mining node's.  Only sending has to be done via an anonymous connection, and as you mentioned earlier, this can be done quite efficiently (I thought transactions were more like 500 bytes, giving a 15x boost to scalability, but whatever.  Actually, as Mike Hearn mentioned somewhere on the forum before, only the first few bytes of transaction hashes actually need to be sent, so this boost can actually be quite a lot larger.).

Unless you mean the anonymous miner's non-anonymized bandwidth is likely smaller than a non-anonymous miner's, since the latter can use an efficient data center.  In which case, how much extra on-demand bandwidth beyond the usual steady transaction flow are you estimating is necessary to thwart these attack blocks?  Is this amount of bandwidth really only available to people who are willing to give up physical control of their machines?  Don't you think that any extra cost due to lowered economic efficiency might be externally funded by parties interested in preventing censorship?

[amincd]

Natural monopolies are...well...natural in some environment.  Namely one's where the barrier to entry is high.  They are also artificially created when possible.  Visa/MC are probably a little bit of both.

I conjecture that efforts based on a freely accessible Bitcoin network will have a very low barrier to entry and can arrange an endless array of unique incentives so I very much doubt that they would fall victim to the natural monopoly effect.

You haven't explained why the tendency for a few dominant networks to arise, which we see in Visa/MC, would not exist with payment processors handling BTC-credit. BTC-credit is not BTC. BTC's decentralization doesn't make credit held by centralized BTC-banks less centralized. The handling of credit happens through traditional off-the-chain channels, where BTC's decentralization has no benefit.

Again with all due respect, it seems like the 'offchain' crew is making assumptions that are not supported by any real substance. The safer assumption, in my opinion, is that centralized parties handling BTC-credit will act like banks that handle other types of credit now, and will be vulnerable to the same type of pressure.

We may have to agree to dis-agree on this, but it is pretty clear to me that we are both handwaving to some extent.

Ultimately, yes.

OK, even if Bitcoin growth happened to stop at a point when a few thousand entities could still operate it...

As I've mentioned on previous threads, there is much more value in having a broad footprint in the network and using it to harvest business intelligence information than there would be in collecting penny-ante fees.  So, entities who have a realistic ability to tap into this source of value will be able to subsidize the services they offer.  This will drive out competition and create a vicious circle of centralization.

A 1 MB block size limit is not a solution to that. Whatever the risks are of having high-bandwidth requirements for nodes, they are nothing compared to the risks that would face unregulated high throughput payment processors.

I'm looking forward to his work along these lines.

Good luck. If you succeed, then you would have no reason to be concerned about a lifting of the 1 MB cap, as off-chain solutions that are as provably secure as the Bitcoin network would be the preferred payment mechanism of users, and most activity will take place off the chain.

[tvbcof]

Natural monopolies are...well...natural in some environment.  Namely one's where the barrier to entry is high.  They are also artificially created when possible.  Visa/MC are probably a little bit of both.

I conjecture that efforts based on a freely accessible Bitcoin network will have a very low barrier to entry and can arrange an endless array of unique incentives so I very much doubt that they would fall victim to the natural monopoly effect.

You haven't explained why the tendency for a few dominant networks to arise, which we see in Visa/MC, would not exist with payment processors handling BTC-credit. BTC-credit is not BTC. BTC's decentralization doesn't make credit held by centralized BTC-banks less centralized. The handling of credit happens through traditional off-the-chain channels, where BTC's decentralization has no benefit.

Again with all due respect, it seems like the 'offchain' crew is making assumptions that are not supported by any real substance. The safer assumption, in my opinion, is that centralized parties handling BTC-credit will act like banks that handle other types of credit now, and will be vulnerable to the same type of pressure.

And as I say, with all due respect, you seem to also.  I reject (fairly vociferously) that there would be a mechanism to promote significant centralization while you assume it would be the case.  I am certain that we won't see the kind of regulatory capture and verndor strong-arming that has put VISA and MC where they are.

Now, it may well be the case that people choose, en-mass, to use one payment processor for the other services they offer (authentication, convenient wallet access, whatever.)  That's lamentable, but it would remain a free choice since any number of alternatives would exist.

I actually think that a very likely outcome would be that people would choose a payment processor who was affiliated with causes they believe in for one reason or another.  Jesuscoin.org, greenpeacecoin.org, nracoin.org, etc.  (The latter will be for sale someday, BTW )

I'll be using some combination of native Bitcoin, some off-chain processor who respects privacy, one who is jurisdictional secure, and probably several who are associated with causes I believe in.  I expect to have a simple UI to effortlessly and inexpensively transfer value between the various off-chain solutions I use.

We may have to agree to dis-agree on this, but it is pretty clear to me that we are both handwaving to some extent.

Ultimately, yes.

Soon.  As soon as we can find it in our hearts to stop stawman-ing one another.

OK, even if Bitcoin growth happened to stop at a point when a few thousand entities could still operate it...

As I've mentioned on previous threads, there is much more value in having a broad footprint in the network and using it to harvest business intelligence information than there would be in collecting penny-ante fees.  So, entities who have a realistic ability to tap into this source of value will be able to subsidize the services they offer.  This will drive out competition and create a vicious circle of centralization.

A 1 MB block size limit is not a solution to that. Whatever the risks are of having high-bandwidth requirements for nodes, they are nothing compared to the risks that would face unregulated high throughput payment processors.

Going full circle, the 1MB (or reasonable) limit is simply to keep the kernel of the distributed crypto-currency ecosystem widely distributed and having a realistic chance of survival under the most significant and coordinated of attacks.

I cannot say that I don't care about the 'spending money' second-tier organs (or 'off-chain processors') but they are in my mind very secondary and very dispensable relative to the actual value core.  They need to compete and offer more options to end-users which I see as a good thing and a much more safe outlet for the urge that everyone feels to have things grow and improve.

[mp420]

Bitcoin is not free, it's about $129 /BTC at the moment. The block size limit won't keep it free but it will surely keep it from rising much above $400, since as we hit the 7tps limit with actual useful (non-SD) transactions the limit will cap the adoption at that level.

Bitcoin is not THAT useful as a technology. Never was, never will be. If we artificially cap the transactions to some level (as opposed to letting the market decide) it will certainly make bitcoin even less useful.

[amincd]

Natural monopolies are...well...natural in some environment.  Namely one's where the barrier to entry is high.  They are also artificially created when possible.  Visa/MC are probably a little bit of both.

I conjecture that efforts based on a freely accessible Bitcoin network will have a very low barrier to entry and can arrange an endless array of unique incentives so I very much doubt that they would fall victim to the natural monopoly effect.

You haven't explained why the tendency for a few dominant networks to arise, which we see in Visa/MC, would not exist with payment processors handling BTC-credit. BTC-credit is not BTC. BTC's decentralization doesn't make credit held by centralized BTC-banks less centralized. The handling of credit happens through traditional off-the-chain channels, where BTC's decentralization has no benefit.

Again with all due respect, it seems like the 'offchain' crew is making assumptions that are not supported by any real substance. The safer assumption, in my opinion, is that centralized parties handling BTC-credit will act like banks that handle other types of credit now, and will be vulnerable to the same type of pressure.

And as I say, with all due respect, you seem to also.  I reject (fairly vociferously) that there would be a mechanism to promote significant centralization while you assume it would be the case.

I believe there would be a mechanism that promotes centralization, because there is a mechanism in traditional payment markets that promotes centralization, and traditional payment processors handle fiat-credit, which has almost exactly the same properties as BTC-credit.

You haven't explained why this mechanism, which we see in fiat payment processing, wouldn't be seen in BTC-credit payment processing, even though they both use the same type of trust-based credit relationships.

Now, it may well be the case that people choose, en-mass, to use one payment processor for the other services they offer (authentication, convenient wallet access, whatever.)  That's lamentable, but it would remain a free choice since any number of alternatives would exist.

It's not a free choice though, because blocks are capped at 1 MB. If there were no block size limit, and they chose to use a centralized payment processor instead of the network, that would be a free choice.

I actually think that a very likely outcome would be that people would choose a payment processor who was affiliated with causes they believe in for one reason or another.  Jesuscoin.org, greenpeacecoin.org, nracoin.org, etc.

This belief rests on the assumption that BTC credit markets would behave differently than fiat credit markets, which you have given no explanation for holding.

I'll be using some combination of native Bitcoin, some off-chain processor who respects privacy, one who is jurisdictional secure, and probably several who are associated with causes I believe in.

Which is no different, from what I can see, in believing that a jurisdictionally secure fiat bank can keep your money secure, and that there would be ubiquitous acceptance of credit from this fiat-bank among merchants. It doesn't happen now, which is why I see no reason it will happen with banks holding BTC-credit.

I expect to have a simple UI to effortlessly and inexpensively transfer value between the various off-chain solutions I use.

The way I see, you might as well expect a simple UI to effortlessly transfer fiat-value between various off-chain solutions. That hundreds of billions of dollars in investment in financial technology has failed to make that a reality thus far, makes me as certain as I could be, that it won't happen with BTC-credit either, because all credit works fundamentally the same way, regardless of the underlying asset that it's based on.

Going full circle, the 1MB (or reasonable) limit is simply to keep the kernel of the distributed crypto-currency ecosystem widely distributed and having a realistic chance of survival under the most significant and coordinated of attacks.

There is very little to attack if there's a 1 MB block size limit. BTC won't be a major player. It'll be centralized banks that handle credit redeemable in BTC that dominate the BTC economy, if indeed BTC becomes widely accepted, which I don't believe it will if transaction fees need to reach $20 to handle global scale traffic.

I'll grant that in the unlikely event that bitcoin succeeded under a 1 MB block cap, and this kind of economy came to be, at least the currency wouldn't be inflated by governments, but everything else will be like the modern financial system, as far as I can see.

I cannot say that I don't care about the 'spending money' second-tier organs (or 'off-chain processors') but they are in my mind very secondary and very dispensable relative to the actual value core.

Which I can't understand. Being able to spend BTC is what gives it value. Low transaction fees give it a competitive advantage. In my opinion, that is what attracts people most to bitcoin.

Bitcoin is not free, it's about $129 /BTC at the moment. The block size limit won't keep it free but it will surely keep it from rising much above $400, since as we hit the 7tps limit with actual useful (non-SD) transactions the limit will cap the adoption at that level.

Bitcoin is not THAT useful as a technology. Never was, never will be. If we artificially cap the transactions to some level (as opposed to letting the market decide) it will certainly make bitcoin even less useful.

That's exactly how I see it.

[Kazimir]

Excellent work on the site and video! I don't fully understand the off-chain solution yet (will read some more on it) but you're doing great work!

[tvbcof]

snip - we are going around in circles on the likely look and feel of BTC-backed second tier processors.  We're not likley to reach a useful conclusion on this so we'll just have to wait and see.

Going full circle, the 1MB (or reasonable) limit is simply to keep the kernel of the distributed crypto-currency ecosystem widely distributed and having a realistic chance of survival under the most significant and coordinated of attacks.

There is very little to attack if there's a 1 MB block size limit. BTC won't be a major player. It'll be centralized banks that handle credit redeemable in BTC that dominate the BTC economy, if indeed BTC becomes widely accepted, which I don't believe it will if transaction fees need to reach $20 to handle global scale traffic.

I'll grant that in the unlikely event that bitcoin succeeded under a 1 MB block cap, and this kind of economy came to be, at least the currency wouldn't be inflated by governments, but everything else will be like the modern financial system, as far as I can see.

There is no limit to the amount of value which could be moved around at 7 transactions per second.

If one could have the same confidence in Bitcoin as they have in gold, it seems likely to me that Bitcoin would end up acting a lot like gold does today on a global level.  While it is downplayed in the West, behind the scenes there huge stockpiles and large (often 'virtual') transfers of it.

In certain societies where there is a cultural affinity (e.g., India), it is common that gold his held as a individual or family store of value.  In other cultures it is not common to do so, but perfectly possible...and has worked out quite well for some.

Two huge advantages that Bitcoin (or something like it) have over gold is that it is inherently auditable and globally fluid.  The latter being important since the barrier to entry for anyone to own it is low.

If real Bitcoin grows to the point where only a relatively small number of highly capitalized entities can operate the system, Bitcoin loses it's value proposition as a reliable store of value.  At least to me and some fraction of the potential user-base.  And if it becomes even mildly popular (by my definition of popular) globally for raw native use, it WILL reach that condition.  In fact, if I were tasked with coming up with a strategy to destroy Bitcoin, this very well may be the approach I would take.

[N12]

First off, I'm not a coder.

As I understand it, the block size limit can only be increased via hard forks which in the future with a much larger userbase and more diverse clients could become infeasible as it would take too long. If the limit is not increased regularly to catch up with increased demand, Bitcoin the block chain will essentially become a system for fat cats who can afford to pay hundreds of today's USD for a single transaction. There would be a divide between the 99% who have no choice but to trust a third party and the 1% who can make use of the block chain's properties. Mainly, no counterparty risk.

Therefore, the solution would have to ensure that the value of a typical transaction's fee stays fairly constant as Bitcoin progresses. Now if there was a way to do it like difficulty, measuring the average fees within a certain amount of blocks and retargeting the block size limit based upon a targeted fee, it would be ideal, but since we will become more reliant on fees given that the block subsidy for miners decreases, I have no idea if this could even be done on paper since that needs to be accounted for, and probably the valuation of Bitcoin would need to be measured too, which is impossible.

I heard about off-chain transactions, but I don't see how it provides non-chain users with 0 counterparty risk. Could users demand some form of bitcoin chain based contract that will release coins eventually if not paid out or withdrawn by the user on a certain date? I doubt it.

Seems like the future will be the lords with the chain and the chained peasants.

[tvbcof]

...
Seems like the future will be the lords with the chain and the chained peasants.

Bitcoin is, by design, a finite resource like gold.

I see no reason why cheap transaction fees will change the reality that the wealthy eventually obtain a fair fraction of the wealth.

In actual fact, it would probably be a lot easier for those with the financial resources to do so, to reach out and grab the BTC if they are flying around at high velocities as Joe Sixpack buys his morning coffee.

Keeping 'native Bitcoin' lite and accessible is mostly a mechanism to keep the actual 'lords with the chain' being pretty much any geek of any nationality situated in any jurisdiction.  Or even floating in the ether between jurisdictions if need be.

[N12]

Bitcoin is, by design, a finite resource like gold.

I see no reason why cheap transaction fees will change the reality that the wealthy eventually obtain a fair fraction of the wealth.

In actual fact, it would probably be a lot easier for those with the financial resources to do so, to reach out and grab the BTC if they are flying around at high velocities as Joe Sixpack buys his morning coffee.

Keeping 'native Bitcoin' lite and accessible is mostly a mechanism to keep the actual 'lords with the chain' being pretty much any geek of any nationality situated in any jurisdiction.  Or even floating in the ether between jurisdictions if need be.

No, naturally Bitcoin changes nothing about inequality of wealth, but that's not what I'm aiming at. The vision of Bitcoin is to introduce digital cash-like transactions to the people rather than having them at risk for confiscated/lost/stolen funds, denied transactions etc.

How ironic would it be if today's "bankster elite" would be some of the only ones to enjoy the properties of Bitcoin all the while serving the 99% via coinbase and bitpay which they control?

[Technomage]

I'm not really interested in debating the issue, since to me it's not an issue. Raising the blocksize limit is the obvious right move since none of the potential concerns (which are not significant) of less centralization due to raising it are worth the limitations it puts on the use of bitcoins, which would in fact make Bitcoin significantly less usable than it is now.

Of all the issues that have been on the table since I started using Bitcoin, I'm most strongly against keeping the blocksize limit as it is. I do understand the issues related to it, all of them, and I'm happy that we're not in a hurry to change it. There is plenty of time to think of the best way to change it. The limit is nowhere near problem zone right now. Eventually however, it will need to be changed.

I hope a super majority of the core Bitcoin community understands this and does not fall to this idiotic idea. The potential advantage of having a maximally decentralized core network only comes into play in extreme situations, while the potential disadvantage of limiting the usability of Bitcoin will come into play immediately, and basically will start making Bitcoin far less useful than it is and thus many users will move on to more scalable networks.

What we need is a scalable and usable payment network, which can be done with Bitcoin. Perhaps not for every microtransaction, but certainly for a lot of things. We do need decentralization as well, but I don't see a problem there either. The concerns over that are highly overrated. Mining and running nodes is already a very specialized operation, and it isn't a problem if it becomes a more specialized task. The network will simply be less decentralized, but that is not the same as centralized.

[tvbcof]

...
How ironic would it be if today's "bankster elite" would be some of the only ones to enjoy the properties of Bitcoin all the while serving the 99% via coinbase and bitpay which they control?

Very.  That is exactly what I see happening as the ability to operate the system progressively squeezes out all but the more well capitalized participants.

One big take-away from the 2013 conference is that most of the devs seem to have very few qualms about introducing 'taint' or 'tarnish' into the system as a means to deter crime.

If this happens at a time when the system is operated almost exclusively by entities who cannot walk away from the investment they have in infrastructure (because they are bound by corporate law if nothing else), I predict that it will be the end of core Bitcoin in anything like what we know it today.  This because the infrastructure operators will be under pressure to leverage the tarnishing system to achieve political goals and will have no choice but to comply.

End users who by necessity, are relying on SPV will have not much say in the matter if there are no SPV servers which they can attach to so they will mutter, but continue to upgrade to software which actually allows them to access their BTC value.

[Technomage]

My first comment was based on the previous discussions over the issue. I decided to check the video, and as usual it has the same fallacies as the previous discussions.

First of all, hardware scaling. I agree that we shouldn't aim for a situation where only large datacenters mine and run nodes, certainly not. That is not what is going to happen though. Even with a 10MB limit and with current hardware, running Bitcoin either as a miner or a full node is not an issue in most places where Bitcoin is popular right now. I for one would have no problem running it with that limit.

A 100MB limit or higher? That's a different story, we can worry about that in the future, when the hardware will most likely be much better, and transferring the blocks optimized etc. Even with today's hardware limiting it to 1MB simply because of centralization fears is absolutely ridiculous. Not that it makes sense to change it right now, it doesn't, but once we hit the limit there is no good reason for not changing it.

The second fallacy is the whole concept of using payment processors for off chain transactions. If we do that, we end up with exactly the same thing as we have now with PayPal etc. I liked the part in the video which explained auditing off chain transactions using cryptography, well that is a great idea.

If that can be viable then I like the idea a lot more, since to me the main reason for using Bitcoin blockchain directly is that it doesn't need to be audited in any way, everything is right there for everyone to verify. I have great distaste for solutions where third party trust is required.

[Technomage]

Also I didn't like the fact that people should lobby against any attempts at raising the limit. Not smart. Most people here are not for blindly raising it, instead a well thought of long term plan must be realized. I will lobby against stupid solutions but in general at this time there is no valid reason for raising it up to 10MB.

Higher than 10MB, I agree that there could be an issue with that. Higher than that is not required in a good while though. When a higher than 10MB limit is actually necessary, it's quite possible that the hardware needed to run the current implementation at the time is actually available for more people than those that have large datacenters.

[ArticMine]

I would suggest that we can support at least 50x the current block size with existing technology and easily keep the ability to run a full node in the hands of many consumers let alone computer enthusiasts. Hard drives in the 3TB are well with the reach of consumers and so are Internet connections that allow 500 GB of data per month. The more important point is that the price of storage, bandwith and computing power is falling faster than any reasonable growth of the Bitcoin network, so while in a few years we could have a block size that approached say 1GB a typical consumer and most certainly an enthusiast would have the computing power, storage and bandwith to be able to handle it.

Let us not forget that 100 years ago sending 1MB of data over the telegraph network would have cost about 50,000 USD in 1913 dollars. Well out of the reach of the average person, but something that J. P. Morgan, the individual, would have been able to afford. Should we limit future generations of Bitcoin users with a hard limit based on today's technology and cost, on the argument that only J. P. Morgan, the corporation, would be able to afford today what in the future would be easily accessible to large numbers of people?

When it comes to the block limit we need a dynamic solution that responds to network demand, difficulty and the real cost of bandwith, digital storage and computing power. If developed correctly this will allow the network to grow while at the same time ensuring that mining and the running of full nodes remains possible for consumers and computer enthusiasts.

By the way transactions outside of the blockchain are not the answer. To understand why one can read http://fincen.gov/statutes_regs/guidance/html/FIN-2013-G001.html.