Economics of block size limit

[oleganza]

I've wrote on my blog about various economic forces behind block size limit. Here's the most important part:

If the miners hit the block limit, it would only mean one thing: there is a desire to process more transactions, but historical untested agreement does not allow it. Then miners will either raise the limit (the smaller the increment, the bigger support it will have), or transaction fees will go up as people compete for the space in blocks. As transaction fees go up, not only miners, but also regular users and service companies using the full blockchain would desire to raise the limit. So it will be easier to make a consensus as demand grows.

My prediction is that the block size limit will probably never be abolished, but will be constantly pushed up by a factor of two as amount of transactions approaches the limit. Maybe after a couple of updates, people would decide that it’s safe to abolish the limit completely if it is cheaper to account for it, than to have uncertainty of a hard fork.

Other concerns are addressed in the full post:
http://blog.oleganza.com/post/43677417318/economics-of-block-size-limit

I did not go deep into analyzing complex schemes like adjustable size limit. Please feel free correct me where I am wrong.

[markm]

Hard forks should also get softer and softer over time.

For example some day one's operating system's package-manager might well be able to check the authenticity of an updated bitcoin package from the source you have designated to it as the one to trust, automatically. Even down to checking whether it is a "critical" / "urgent" update or a mere non-critical bugfix update.

The software could even come with a "warning, this version should not be relied upon to remain up to date beyond (some date about a year after its release)".

Any time bitcoin exchange rates fall drastically in response to too many blocks being full of paying transactions too much of the day/night, machines all over the world could find, during their nightly automatic check for package updates, an update ready for them to automatically install.

Sure a lot of security and trust details to get that all down pat, but provably repeatable builds from reputably attestedly audited and tested source code, which is already in place, is a nice step already in that direction.

-MarkM-

[gmaxwell]

I've wrote on my blog about various economic forces behind block size limit. Here's the most important part:

If the miners hit the block limit, it would only mean one thing: there is a desire to process more transactions, but historical untested agreement does not allow it. Then miners will either raise the limit

I did not go deep into analyzing complex schemes like adjustable size limit. Please feel free correct me where I am wrong.

You are, apparently, blogging about technology you do not understand. Please stop. Miners cannot "raise the limit"— that isn't how Bitcoin works. That precisely the same as saying that when the subsidy halved from 50 to 25 miners would continue on mining 50 because they prefer the greater income. They may have wanted to— but the system doesn't permit it, and the miners can't change the system. Darn good that they can't because otherwise all our expectations about the rules that make Bitcoin valuable would be worthless— subject to the whim of a small pool of anonymous interests.

[MoonShadow]

Any time bitcoin exchange rates fall drastically in response to too many blocks being full of paying transactions

Honestly, Mark.  Do you not see the contradiction in the above statement?  What makes you believe that the rate of transactions are, in any way, related to the exchange rates?

[markm]

Any time bitcoin exchange rates fall drastically in response to too many blocks being full of paying transactions

Honestly, Mark.  Do you not see the contradiction in the above statement?  What makes you believe that the rate of transactions are, in any way, related to the exchange rates?

The FUD brigade's sky-is-falling / sky-will-fall FUD about how failing to make blocks too big too soon will cause a mass migration of users, hence thus presumably also value (else they aren't the important users anyway, maybe) from bitcoin and/or a failure to bring in new users (which, combined with the inflation rate from minting, also could lower exchange rates).

Or in short, "if exchange rates aren't falling, the sky ain't either".

-MarkM-

[markm]

...

Maybe a little harsh, there; my take on the OP was more toward responding "yes, it is a market in action, if it pays everyone it will happen".

As it didn't seem to me to be saying miners could force it up by mining so much as that the soft limits we have in place that miners do get to comment out or adjust to their liking will provide feedback.

In other words, knowing that the miners only have the soft limits to adjust, benefit of the doubt along with recognition that the whole hard versus soft limits techie details was maybe a bit rich for his intended audience led me to the idea that speaking in terms of the soft limits rather than the hard limit possibly suited the article better.

-MarkM-

[MoonShadow]

Or in short, "if exchange rates aren't falling, the sky ain't either".

I can accept that.

[oleganza]

If the miners hit the block limit, it would only mean one thing: there is a desire to process more transactions, but historical untested agreement does not allow it. Then miners will either raise the limit

I did not go deep into analyzing complex schemes like adjustable size limit. Please feel free correct me where I am wrong.

You are, apparently, blogging about technology you do not understand. Please stop. Miners cannot "raise the limit"— that isn't how Bitcoin works. That precisely the same as saying that when the subsidy halved from 50 to 25 miners would continue on mining 50 because they prefer the greater income. They may have wanted to— but the system doesn't permit it, and the miners can't change the system. Darn good that they can't because otherwise all our expectations about the rules that make Bitcoin valuable would be worthless— subject to the whim of a small pool of anonymous interests.

Thanks for reply. I didn't mean unilateral change by miners only. I meant consensus that must be achieved by everyone who runs the full node. I have rephrased the sentence to avoid the confusion.

Is there anything else incorrect or badly written?

[jgarzik]

It is crucial to understand the concept and, yes, economic impact of a hard fork before even approaching the economic analysis of changing the max block size.

A hard fork is a significant event that knocks legitimate users off the network, makes coins unspendable, or potentially makes the same coins spendable in two different locations, depending on whether or not you're talking to an updated node.

It is, to pick a dramatic term, an Extinction Level Event.  If done poorly, a hard fork could make it impossible for reasonable merchants to trust the bitcoins they receive, the very foundation of their economic value.

Furthermore, a hard fork is akin to a Constitutional Convention:  a hard fork implies the ability to rewrite the ground rules of bitcoin, be it block size, 21M limit, SHA256 hash, or other hard-baked behavior.  Thus, there is always the risk of unpredictable miners, users and devs changing more than just the block size precisely because it makes the most engineering sense to change other hard-to-change features at the time of hard-fork.

It is a nuclear option with widespread economic consequences for all bitcoin users.

[Zangelbert Bingledack]

I've wrote on my blog about various economic forces behind block size limit. Here's the most important part:

If the miners hit the block limit, it would only mean one thing: there is a desire to process more transactions, but historical untested agreement does not allow it. Then miners will either raise the limit (the smaller the increment, the bigger support it will have), or transaction fees will go up as people compete for the space in blocks. As transaction fees go up, not only miners, but also regular users and service companies using the full blockchain would desire to raise the limit. So it will be easier to make a consensus as demand grows.

My prediction is that the block size limit will probably never be abolished, but will be constantly pushed up by a factor of two as amount of transactions approaches the limit. Maybe after a couple of updates, people would decide that it’s safe to abolish the limit completely if it is cheaper to account for it, than to have uncertainty of a hard fork.

Other concerns are addressed in the full post:
http://blog.oleganza.com/post/43677417318/economics-of-block-size-limit

I did not go deep into analyzing complex schemes like adjustable size limit. Please feel free correct me where I am wrong.

Finally, common sense. Static thinking and imagining people are robots who won't change their behavior dynamically in response to change is rife, maybe because coders are not used to incorporating human action into their considerations. https://www.youtube.com/watch?v=BbGKkV-WoYg

[oleganza]

I wouldn't be so dramatic about hard fork. People will not change their software unless they are 99,99% sure everyone else is going to do the same. If there is no way to "vote", then nothing will change. If block size limit or some other rule will get in the way of many users, everyone's desire to find a good enough mechanism of voting will get stronger until it is finally found, implemented and the change is applied.

No miner, user or developer is going to "unpredictably" change any rule if he is not sure others are doing the same. There is no nuclear situation at all. If someone is not smart enough to update without a proof of consensus, he and his customers will suffer. But not the whole network. Thankfully, there are plenty of businesses, nodes and miners, so stupidity of one of them will not significantly affect trust in the network.

I guess, we already have a way to vote for an new proposal using block version numbers: https://en.bitcoin.it/wiki/BIP_0034

Say, "block version 3 increases limit to 2 Mb". Miners who agree with that will upgrade to version 3, but still keeping the block size under 1 Mb. If > 90% of the blocks is version 3, it will show the consensus. Those who disagree (including non-mining nodes) will have data for calculating their risks. We can see that 90% consensus is possible only if 90% of miners really believe they have a lot of valuable transactions, not spam, and that it's profitable for them to increase the limit. If only a couple of miners want to increase the limit, there will be no consensus.

When the miners vote, the natural question is whether miners have opposite interests to those of other full nodes. I think the interests are aligned in case of a block size limit. Every node that creates transactions, competes with other nodes. And the more transactions the node creates, the more expensive it is going to be (think bitpay, coinbase etc.), hence the bigger incentive for them to accept larger blocks. When they see majority of miners accepting new rule, they are going to accept it too with probability proportional to their transaction expenses (which is proportional to number of transactions). In the end, the opposition to miner's consensus will consist of only nodes with little originating transactions. And these remaining non-mining nodes may choose not propagate new blocks, but it does not matter as there are a lot of other nodes who will. Disagreeing minority of miners will of course prefer certainty of block rewards to the risk of producing orphaned blocks and will accept the new rule.

Last point: if the suggested block size limit is too high, people would have more reservations against it. If, however, it is only 2x the current limit, the cost of bandwidth and storage is much easier to calculate, so everyone would feel good enough. And if it works out well for everybody, we can do another round of increase when new limit is filled up. If it does not work well for everybody, miners will simply create smaller blocks.

In the end, I don't think anybody would bother upgrading until we start approaching 1 Mb limit for real. When we will, it will be much clear for both miners and non-miners how much the change is desired.

[markm]

(Re the post above...)

Very well written, thank you.

Not sure folk are going to be able  to come up with much to refute it, either.

Maybe we should hope you aren't right simply because other people being right is one of the most annoying things in the whole wide world thus could lead to flame wars or something.

-MarkM-

[TierNolan]

Say, "block version 3 increases limit to 2 Mb". Miners who agree with that will upgrade to version 3, but still keeping the block size under 1 Mb. If > 90% of the blocks is version 3, it will show the consensus. Those who disagree (including non-mining nodes) will have data for calculating their risks. We can see that 90% consensus is possible only if 90% of miners really believe they have a lot of valuable transactions, not spam, and that it's profitable for them to increase the limit. If only a couple of miners want to increase the limit, there will be no consensus.

It might also be worth adding an delay to activation.  If 90% of blocks between 235000 and 250000 are version 3, then starting with block 270000, the block size will be 2MB.

I think the interests are aligned in case of a block size limit. Every node that creates transactions, competes with other nodes.

Maintaining independent verification is important too.

There is probably a curve for optimal profits.  If the block size is infinite, then there are no fees due to space requirements.  Similarly, if there is no space, fees are zero due to no transactions.

Like the Laffer curve, there would be an optimal (to miners) size, which draws in the maximum amount of fees.

If miners, vote, then they will aim for that.

[twolifeinexile]

Furthermore, a hard fork is akin to a Constitutional Convention:  a hard fork implies the ability to rewrite the ground rules of bitcoin, be it block size, 21M limit, SHA256 hash, or other hard-baked behavior.  Thus, there is always the risk of unpredictable miners, users and devs changing more than just the block size precisely because it makes the most engineering sense to change other hard-to-change features at the time of hard-fork.
It is a nuclear option with widespread economic consequences for all bitcoin users.

I would repeat this.

[nevafuse]

I couldn't agree with your analysis more.  I personally think the limit should be abolished along with a few other constraints.  Although I'm not sure the community agrees as a whole.  The exchanges have the most say for now because they are most centralized part of bitcoin.  If mtgox decides to fork, then the whole community will.  Although to be honest, bitcoin won't last forever.  We'll get to a point where forking is too difficult & switching to another cryptocurrency is an easier option.  And it may make more sense to forget forking at all because bitcoin is too fragile & I'd rather at least one cryptocurrency succeed for now despite its flaws (which pale in comparison with non-cryptocurrencies).

[MoonShadow]

Furthermore, a hard fork is akin to a Constitutional Convention:  a hard fork implies the ability to rewrite the ground rules of bitcoin, be it block size, 21M limit, SHA256 hash, or other hard-baked behavior.  Thus, there is always the risk of unpredictable miners, users and devs changing more than just the block size precisely because it makes the most engineering sense to change other hard-to-change features at the time of hard-fork.
It is a nuclear option with widespread economic consequences for all bitcoin users.

I would repeat this.

While I do agree, this is also an argument for getting things settled sooner rather than later. 

[conv3rsion]

It is crucial to understand the concept and, yes, economic impact of a hard fork before even approaching the economic analysis of changing the max block size.

A hard fork is a significant event that knocks legitimate users off the network, makes coins unspendable, or potentially makes the same coins spendable in two different locations, depending on whether or not you're talking to an updated node.

It is, to pick a dramatic term, an Extinction Level Event.  If done poorly, a hard fork could make it impossible for reasonable merchants to trust the bitcoins they receive, the very foundation of their economic value.

Furthermore, a hard fork is akin to a Constitutional Convention:  a hard fork implies the ability to rewrite the ground rules of bitcoin, be it block size, 21M limit, SHA256 hash, or other hard-baked behavior.  Thus, there is always the risk of unpredictable miners, users and devs changing more than just the block size precisely because it makes the most engineering sense to change other hard-to-change features at the time of hard-fork.

It is a nuclear option with widespread economic consequences for all bitcoin users.

That puts it in very good perspective. Thank you for that post.

[johnyj]

After a hard fork, even there are only 4 people hashing on the original chain with old client, it will quickly develop into a popular chain and most possibly knock out the new chain

It is a no brainer: One chain has limited resource (expensive), another chain has abundant resource (cheap), btc on which chain will worth more?

[twolifeinexile]

After a hard fork, even there are only 4 people hashing on the original chain with old client, it will quickly develop into a popular chain and most possibly knock out the new chain

It is a no brainer: One chain has limited resource (expensive), another chain has abundant resource (cheap), btc on which chain will worth more?

Not necessarily,people also need to consider long term success of which fork.

[johnyj]

After a hard fork, even there are only 4 people hashing on the original chain with old client, it will quickly develop into a popular chain and most possibly knock out the new chain

It is a no brainer: One chain has limited resource (expensive), another chain has abundant resource (cheap), btc on which chain will worth more?

Not necessarily,people also need to consider long term success of which fork.

Gold seldom get moved more than once a year
Fiat money moves thousands of times a second around the world

Remove the block size limit won't make the scalability any better, since then you have exponential growth of bandwidth and disk space, and fall into the exactly the trajectory as today's monetary system: If bandwidth and disk space did not grow, bitcoin will die, in order for bitcoin to survive, the internet must grow...