BIP1xx DynamicMaxBlockSize

[adamstgBit]

why is nobody talking about BIP1xx
https://github.com/UpalChakraborty/bips/blob/master/BIP-DynamicMaxBlockSize.mediawiki

who knew there could be so many possible solutions to scaling bitcoin 

[Lauda]

I'm pretty sure that if we keep this up we will never reach an agreement. Why not just somehow implement the calculating from this proposal into BIP100 (remove the miners voting) and we could all be happy.
Then BIP100 would sound much better than and we would already have something that people are aware of.

I'd like to see some comments from the developers though.

[Klestin]

why is nobody talking about BIP1xx
https://github.com/UpalChakraborty/bips/blob/master/BIP-DynamicMaxBlockSize.mediawiki

who knew there could be so many possible solutions to scaling bitcoin 

Because:

- Scaling based on previous block size allows a single motivated attacker to spam the block chain so that all blocks are full.  This would result in a doubling of the block size cap.  Rinse, repeat.  The net result is that the block size cap is effectively non-existant.  If you think BIP 101 grows too fast, BIP 1xx can be forced to grow at ludicrous speed.

- Depending on previous block size and transaction fee would mitigate this. HOWEVER: That's as far as they've gotten with the idea. There is no actual proposal that factors fees.

[CounterEntropy]

- Scaling based on previous block size allows a single motivated attacker to spam the block chain so that all blocks are full.  This would result in a doubling of the block size cap.  Rinse, repeat.  The net result is that the block size cap is effectively non-existant.  If you think BIP 101 grows too fast, BIP 1xx can be forced to grow at ludicrous speed.

To keep the Max cap increasing, attacker needs to keep burning his coins. And once he burns out, the cap will automatically scale down. In reality, an attacker, who is willing to burn his coin can always play with bitcoin network. Recent spam experiments prove that beyond doubt. On the other hand, BIP 101 only monotnously increase the max cap, without accounting for the market demand. It does not adjust the max cap with decreasing demand for space in blocks.

- Depending on previous block size and transaction fee would mitigate this. HOWEVER: That's as far as they've gotten with the idea. There is no actual proposal that factors fees.

This allows a market for Tx fee bidding, so that miners be subsidized when block reward goes down. Parameters may be fine tuned. But, BIP 1xx is the best proposal in my opinion.

[adamstgBit]

its too bad BIP1xx isnt getting any love its so good.

[CounterEntropy]

its too bad BIP1xx isnt getting any love its so good.

How would it ? Gregory Maxwell aka BlockStream is not assigning a BIP no. to it, though it is long been discussed on the bitcoin-dev list.

http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-August/subject.html#10285

[RustyNomad]

I like the dynamics of this proposal.

Its important that the blocksize is able to adjust both up and down. As mentioned above, a spammer can increase it yes but with each step it would become more and more expensive to maintain and once the 'attack' is over the block sizes scale down again in line with the actual demand of the network.

We do not know what the demand will be like a month or a year or even a decade from now so the best solution is one that's based on actual usage. Large scale adoption could happen next year or it might only happen in 5 years time. Working on a model that just increases the block size based on assumptions of what the actual demand will be at a certain point in time is in my opinion flawed and will bring us right back to where we are now discussing block sizes.

[Klestin]

To keep the Max cap increasing, attacker needs to keep burning his coins. And once he burns out, the cap will automatically scale down. In reality, an attacker, who is willing to burn his coin can always play with bitcoin network. Recent spam experiments prove that beyond doubt. On the other hand, BIP 101 only monotnously increase the max cap, without accounting for the market demand. It does not adjust the max cap with decreasing demand for space in blocks.

The exact same cost is incurred if ANY spammer wants to fill blocks.  Anyone willing to accept that as a safety net should have zero problem with the exponential growth in BIP 101, since that increase in maximum block size will have zero effect on the actual block size unless they're filled.  If they're being filled by commerce, great!  If they're being filled by spammers, then at least there is a hard cap with BIP 101.  BIP 1xx does not have the safety of that hard cap.

BIP 1xx as written also has the reverse problem.  A few miners can collude to restrict transaction volume in their blocks, causing a repeated halving of block size CAP.  A rational minority miner who would otherwise be willing to include all those transactions will find his block size restricting repeatedly.

I realize that these two scenarios are mutually exclusive - they can't both happen simultaneously.  Further, I would agree that neither of them is particularly likely.  The problem I have is that I don't see a rational reason to accept these possibilities.  BIP-101 does not allow for either of these (unlikely, but terrible) outcomes.  It follows a predictive curve, just like block rewards.

[tvbcof]

I'm pretty sure that if we keep this up we will never reach an agreement. Why not just somehow ...

We are never going to reach an agreement.  The sides are way to far apart.

Just do XT/BIP101 and fork the thing.  If people want to try a dynamic adjustment and cannot meet eye-to-eye with the Hearnian people or small-blockists, fork three ways.  Then each side goes on their merry way and makes the best of the basic strategy they've chosen.

It's not even relevant to say 'let the best man win.'  There are advantages and disadvantages to each strategy.  Forking the blockchain is the best way to realize the advantages of each.

I firmly believe that this strategy is the best way to move Satoshi's ideas (whatever they may be) forward, and while there might be some argument that it would damage certain things about Bitcoin they are philosophically fairly weak and the damage being done by trying to make dogs and cats live together in harmony will be far worse.

[adamstgBit]

I'm pretty sure that if we keep this up we will never reach an agreement. Why not just somehow ...

We are never going to reach an agreement.  The sides are way to far apart.

Just do XT/BIP101 and fork the thing.  If people want to try a dynamic adjustment and cannot meet eye-to-eye with the Hearnian people or small-blockists, fork three ways.  Then each side goes on their merry way and makes the best of the basic strategy they've chosen.

It's not even relevant to say 'let the best man win.'  There are advantages and disadvantages to each strategy.  Forking the blockchain is the best way to realize the advantages of each.

I firmly believe that this strategy is the best way to move Satoshi's ideas (whatever they may be) forward, and while there might be some argument that it would damage certain things about Bitcoin they are philosophically fairly weak and the damage being done by trying to make dogs and cats live together in harmony will be far worse.

merchants would be even more confused than they are today about how to accept "Bitcoin", if we split bitcoin now bitcoin dies.

what is bitcoin? well which one? there are 3 and they're all exactly the same expect there treament of blocklimit.

that'll destroy bitcoin.

[achow101]

To keep the Max cap increasing, attacker needs to keep burning his coins. And once he burns out, the cap will automatically scale down. In reality, an attacker, who is willing to burn his coin can always play with bitcoin network. Recent spam experiments prove that beyond doubt. On the other hand, BIP 101 only monotnously increase the max cap, without accounting for the market demand. It does not adjust the max cap with decreasing demand for space in blocks.

The exact same cost is incurred if ANY spammer wants to fill blocks.  Anyone willing to accept that as a safety net should have zero problem with the exponential growth in BIP 101, since that increase in maximum block size will have zero effect on the actual block size unless they're filled.  If they're being filled by commerce, great!  If they're being filled by spammers, then at least there is a hard cap with BIP 101.  BIP 1xx does not have the safety of that hard cap.

With BIP 101, spammers can continue to spam blocks and know for certain that for a few years they can both start another debate and the cost of continuous spam does not change that much. With BIP 1xx, the cost of spamming full blocks will double every two weeks, thus making it incredibly expensive to maintain a prolonged spam attack on Bitcoin. With the same amount of money, a spam attack would last a shorter amount of time if BIP 1xx was implemented than with BIP 101.

BIP 1xx as written also has the reverse problem.  A few miners can collude to restrict transaction volume in their blocks, causing a repeated halving of block size CAP.  A rational minority miner who would otherwise be willing to include all those transactions will find his block size restricting repeatedly.

I realize that these two scenarios are mutually exclusive - they can't both happen simultaneously.  Further, I would agree that neither of them is particularly likely.  The problem I have is that I don't see a rational reason to accept these possibilities.  BIP-101 does not allow for either of these (unlikely, but terrible) outcomes.  It follows a predictive curve, just like block rewards.

BIP 101 allows for a spammer to continuously fill blocks for a prolonged period of time, and thus sparking another debate like this on about how the block size limit is too small. It can cause a scenario like what we have now.

[tvbcof]

We are never going to reach an agreement.  The sides are way to far apart.

Just do XT/BIP101 and fork the thing.  If people want to try a dynamic adjustment and cannot meet eye-to-eye with the Hearnian people or small-blockists, fork three ways.  Then each side goes on their merry way and makes the best of the basic strategy they've chosen.

It's not even relevant to say 'let the best man win.'  There are advantages and disadvantages to each strategy.  Forking the blockchain is the best way to realize the advantages of each.

I firmly believe that this strategy is the best way to move Satoshi's ideas (whatever they may be) forward, and while there might be some argument that it would damage certain things about Bitcoin they are philosophically fairly weak and the damage being done by trying to make dogs and cats live together in harmony will be far worse.

merchants would be even more confused than they are today about how to accept "Bitcoin", if we split bitcoin now bitcoin dies.

what is bitcoin? well which one? there are 3 and they're all exactly the same expect there treament of blocklimit.

that'll destroy bitcoin.

That's what I mean by 'philosophically fairly weak'.

There should be no problem with merchants or anyone else either accepting or not accepting certain flavors.  Having a USD, a EURO, a Yuan, etc has not killed commerce.  If such a trivial thing is so difficult for them they would not be able to run a business in the first place, and from the consumer end, anyone who could not figure it out probably belongs in a nursing home.

Anyway, merchants decisions will likely be made for them when the obtain a crypto-currency license from the state.

At some point you need to come to terms with the fact that making a baloney argument then following it up with a declaration like "that'll destroy bitcoin." is simply not a strategy which works on everyone.  Probably not even a majority of people, and right about zero percentage of thinking people who matter.

[adamstgBit]

We are never going to reach an agreement.  The sides are way to far apart.

Just do XT/BIP101 and fork the thing.  If people want to try a dynamic adjustment and cannot meet eye-to-eye with the Hearnian people or small-blockists, fork three ways.  Then each side goes on their merry way and makes the best of the basic strategy they've chosen.

It's not even relevant to say 'let the best man win.'  There are advantages and disadvantages to each strategy.  Forking the blockchain is the best way to realize the advantages of each.

I firmly believe that this strategy is the best way to move Satoshi's ideas (whatever they may be) forward, and while there might be some argument that it would damage certain things about Bitcoin they are philosophically fairly weak and the damage being done by trying to make dogs and cats live together in harmony will be far worse.

merchants would be even more confused than they are today about how to accept "Bitcoin", if we split bitcoin now bitcoin dies.

what is bitcoin? well which one? there are 3 and they're all exactly the same expect there treament of blocklimit.

that'll destroy bitcoin.

That's what I mean by 'philosophically fairly weak'.

There should be no problem with merchants or anyone else either accepting or not accepting certain flavors.  Having a USD, a EURO, a Yuan, etc has not killed commerce.  If such a trivial thing is so difficult for them they would not be able to run a business in the first place, and from the consumer end, anyone who could not figure it out probably belongs in a nursing home.

Anyway, merchants decisions will likely be made for them when the obtain a crypto-currency license from the state.

At some point you need to come to terms with the fact that making a baloney argument then following it up with a declaration like "that'll destroy bitcoin." is simply not a strategy which works on everyone.  Probably not even a majority of people, and right about zero percentage of thinking people who matter.

LOL

wtv man try it and watch everyone back away from bitcoin and then run like hell.

[adamstgBit]

We are never going to reach an agreement.  The sides are way to far apart.

Just do XT/BIP101 and fork the thing.  If people want to try a dynamic adjustment and cannot meet eye-to-eye with the Hearnian people or small-blockists, fork three ways.  Then each side goes on their merry way and makes the best of the basic strategy they've chosen.

It's not even relevant to say 'let the best man win.'  There are advantages and disadvantages to each strategy.  Forking the blockchain is the best way to realize the advantages of each.

I firmly believe that this strategy is the best way to move Satoshi's ideas (whatever they may be) forward, and while there might be some argument that it would damage certain things about Bitcoin they are philosophically fairly weak and the damage being done by trying to make dogs and cats live together in harmony will be far worse.

merchants would be even more confused than they are today about how to accept "Bitcoin", if we split bitcoin now bitcoin dies.

what is bitcoin? well which one? there are 3 and they're all exactly the same expect there treament of blocklimit.

that'll destroy bitcoin.

That's what I mean by 'philosophically fairly weak'.

There should be no problem with merchants or anyone else either accepting or not accepting certain flavors.  Having a USD, a EURO, a Yuan, etc has not killed commerce.  If such a trivial thing is so difficult for them they would not be able to run a business in the first place, and from the consumer end, anyone who could not figure it out probably belongs in a nursing home.

Anyway, merchants decisions will likely be made for them when the obtain a crypto-currency license from the state.

At some point you need to come to terms with the fact that making a baloney argument then following it up with a declaration like "that'll destroy bitcoin." is simply not a strategy which works on everyone.  Probably not even a majority of people, and right about zero percentage of thinking people who matter.

LOL

wtv man try it and watch everyone back away from bitcoin and then run like hell.

its not a bad idea to have different implementations of bitcoin competing, i think the alt coins do just that, and splitting bitcoin over this tiny detail is going to do lot more harm then good.

[RocketSingh]

its not a bad idea to have different implementations of bitcoin competing, i think the alt coins do just that, and splitting bitcoin over this tiny detail is going to do lot more harm then good.

Bitcoin is not like linux, which can have different flavor. It is a medium of exchange and a medium of exchange requires trust. If bitcoin splits, that would severely hamper that trust.

[tvbcof]

its not a bad idea to have different implementations of bitcoin competing, i think the alt coins do just that, and splitting bitcoin over this tiny detail is going to do lot more harm then good.

It's not a 'tiny detail' to me in the slightest.  It cuts to the very heart of what distributed crypto-currencies are and what they are realistically good for.

Bitcoin is not like linux, which can have different flavor. It is a medium of exchange and a medium of exchange requires trust. If bitcoin splits, that would severely hamper that trust.

That is simply not true in my case.  What is 'hampering trust' in my mind is the 'Free Shit Nation' gaggle of hangers-on who demand free transactions that are subsidized by others because that is how life has always worked for them.  I would love to wave them goodbye and wish them the best on their bloatchain fork then be able to move forward building a rock solid and truly distributed foundation which would have a multitude of uses.  Including, ironically, many opportunities for the 'Free Shit Minions' to get exactly what they are after.

[RocketSingh]

Bitcoin is not like linux, which can have different flavor. It is a medium of exchange and a medium of exchange requires trust. If bitcoin splits, that would severely hamper that trust.

That is simply not true in my case.  What is 'hampering trust' in my mind is the 'Free Shit Nation' gaggle of hangers-on who demand free transactions that are subsidized by others because that is how life has always worked for them.  I would love to wave them goodbye and wish them the best on their bloatchain fork then be able to move forward building a rock solid and truly distributed foundation which would have a multitude of uses.  Including, ironically, many opportunities for the 'Free Shit Minions' to get exactly what they are after.

Seems you support the bidding of Tx fee to subsidize miners while block subsidy goes down. In that case, Proposal 2 of BIP 1xx might interest you.

Personally, I am undecided about this. Because one of the reason many people got attracted towards bitcoin was very cheap or almost no transfer cost. For those who thinks the same may like Proposal 1 of BIP 1xx.

[unamis76]

Talks about BIP's are finally converging, and scaling is also a nice option... I voted no preference anyways. I think BIP 100 and 101 have a bit to learn about each other The virtue is in the middle. And maybe for the future we can think of scaling solutions. Or even now! Who knows... But scaling has some visible flaws as seen on this thread Also, BIP100 allows enough scalability...

[Klestin]

With BIP 101, spammers can continue to spam blocks and know for certain that for a few years they can both start another debate and the cost of continuous spam does not change that much. With BIP 1xx, the cost of spamming full blocks will double every two weeks, thus making it incredibly expensive to maintain a prolonged spam attack on Bitcoin. With the same amount of money, a spam attack would last a shorter amount of time if BIP 1xx was implemented than with BIP 101.

It sounds like you have the idea of spamming blocks backwards.  Spammers can't block legitimate transactions - the best they can hope is to delay no-fee transactions.  A block isn't filled with the first transactions that a miner sees - the miner decides which transactions to include, based on the transaction size/priority/fee.

This is not what the block size cap is there to combat.  The anti-spam limit of 1MB is there to prevent a spammer from filling the hard drives of every node operator.  This is why we're not discussing moving to unlimited block size.

With BIP 101, there is no more incentive for spammers to attack than there is right now - and there has never been a spam attack in the history of Bitcoin.

With BIP 1xx, there is absolutely something that a motivated spammer could do - they could directly control the cap, moving it higher and higher.  As I said previously though, that's not really what I worry about.  I worry about the reverse.  Miners can truncate blocks, and directly control how much other miners can include in their blocks.  It allows crippling of the open-market transaction processing that we now have.  Why abdicate this to the miners - a group that has previously advocated for smaller blocks with the stated rationale that it may force higher transaction fees?  It is the fox guarding the henhouse.

And no, miners can't do this right now.  Yes, they can control their own blocks, but all it takes is any one miner to allow those other transactions in with small/no fee.  This is a check against the monopoly that could result from BIP 1xx.

BIP 1xx is union rules for miners.  Those rogue miners who will process all your low-fee transactions?  They could be branded scabs, and the union masters could shrink the block size so far that the transaction fees are forced higher regardless of the remaining miners.

[achow101]

With BIP 101, spammers can continue to spam blocks and know for certain that for a few years they can both start another debate and the cost of continuous spam does not change that much. With BIP 1xx, the cost of spamming full blocks will double every two weeks, thus making it incredibly expensive to maintain a prolonged spam attack on Bitcoin. With the same amount of money, a spam attack would last a shorter amount of time if BIP 1xx was implemented than with BIP 101.

It sounds like you have the idea of spamming blocks backwards.  Spammers can't block legitimate transactions - the best they can hope is to delay no-fee transactions.  A block isn't filled with the first transactions that a miner sees - the miner decides which transactions to include, based on the transaction size/priority/fee.

Those spam transactions can delay low fee transactions for a long time.

This is not what the block size cap is there to combat.  The anti-spam limit of 1MB is there to prevent a spammer from filling the hard drives of every node operator.  This is why we're not discussing moving to unlimited block size.

With BIP 101, there is no more incentive for spammers to attack than there is right now - and there has never been a spam attack in the history of Bitcoin.

Where were you a month ago when someone was spamming transactions? Don't give me BS about a "stress test" it was most definitely a spam attack. It caused 80000 unconfirmed txs, many of them legitimate.

With BIP 1xx, there is absolutely something that a motivated spammer could do - they could directly control the cap, moving it higher and higher.  As I said previously though, that's not really what I worry about.  I worry about the reverse.  Miners can truncate blocks, and directly control how much other miners can include in their blocks.  It allows crippling of the open-market transaction processing that we now have.  Why abdicate this to the miners - a group that has previously advocated for smaller blocks with the stated rationale that it may force higher transaction fees?  It is the fox guarding the henhouse.

And no, miners can't do this right now.  Yes, they can control their own blocks, but all it takes is any one miner to allow those other transactions in with small/no fee.  This is a check against the monopoly that could result from BIP 1xx.

BIP 1xx is union rules for miners.  Those rogue miners who will process all your low-fee transactions?  They could be branded scabs, and the union masters could shrink the block size so far that the transaction fees are forced higher regardless of the remaining miners.

While that is possible, you need to see it the way that miners see it and do some cost-benefit analysis.

Preventing transactions could cause several negative outcomes for miners. They would be losing out on transaction fees if a fee market doesn't develop. If the blocksize goes too low or fees go too high, people might just stop using Bitcoin causing the price to drop and thus miners are losing money. An increased backlog of transactions can negatively impact the full node by filling its mempool to the point that the node just crashes.

The only positive that could come out of causing small blocksizes would be the possibility of a fee market, which compared to the risks, is not something they would attempt with their already small profit margins

Also, if you think that miners would do this, then why are so many of them voting for 8 MB blocks?

A solution to this problem you present would be to create a lower bound on the blocksize limit. It would be that the maximum blocksize cannot go any lower than say 1 MB (or any other number you like).