Bitcoin 20MB Fork

[DooMAD]

@RoadStress
I'd feel like using something else to move the value.
I'm quite positive I wouldn't feel like "omg the BTC dream has ended".



I really don't get it. You guys in the pro side help me understand your position.
Because for me the choice is very clear:

1) We don't necessarily abandon the microtrasactions (trying to develop them with sidechains) AND keep the security and decentralization;
OR
2) We embrace the microtransations AND centralize hoping this wouldn't be a security issue.


For now I have no doubt in being anti-fork.

The question you have to ask yourself is, will you ever actually get sidechains if you stay on the 1MB chain?  The anti-fork crowd have gone on record as saying they'd like to set up a bunch of nodes running the current version that can be left running and basically forgotten about, so if there are new features people want to see added, you'll be out of luck if the network isn't updated to support that new feature.  

People might be paranoid about the supposed "threat" from the new fork,  but at least that new fork will be prepared to adapt and make changes if the situation calls for it.  It seems like the anti-fork crowd just want to freeze this present moment in time as the pinnacle of Bitcoin's development and leave it like it is now for the remainder of time.  Yet another reason why you're going to get left behind if you stay on the old 1MB limited chain.  If they don't innovate, they won't stay relevant.

[amincd]

Forgive me for being stupid but does this mean that the maximum block size will increase automatically? 40% every year and then it will stop after reaching 20MB^10?
I'm not good at math...

It will stop after doubling 10 times, meaning at 16.8 MB * 2^10. The maximum maximum block size will become 16 GB on Jan 1 2035. It won't increase any more after that.

[spazzdla]

If the blockchain gets ~ 2 000 000 000 GB Bitcoin WILL FAIL.

[hdbuck]

Have we tried increasing the block size limit on a test coin yet?  We should make sure the functions operate as expected under high and bursts of saturation workloads.

Here: http://gavintech.blogspot.com/2015/01/twenty-megabytes-testing-results.html

Why exactly is this thread titled "20 MB fork" when what he is really talking about is :
- 25 MB blocks next year
- 35 MB blocks in two years
- 100 MB blocks in 5 years
- ~500 MB blocks in 10 years

Ask gavin.

[Cryddit]

The question you have to ask yourself is, will you ever actually get sidechains if you stay on the 1MB chain?  The anti-fork crowd have gone on record as saying they'd like to set up a bunch of nodes running the current version that can be left running and basically forgotten about, so if there are new features people want to see added, you'll be out of luck if the network isn't updated to support that new feature.  

No.  You won't get side-chains if you stay on the 1MB block chain, because the 1MB block chain will not move fast enough to support them.  If anyone at all stays with a 1MB block chain after the fork, then either they will have to have another hard fork to reset the difficulty, or they'll have to adapt and cope with a chain that gets a block less often than once per day.  It would be at least a decade and more probably a lifetime before that network recovered to the extent of being able to handle even one transaction per second. 

It will be quite amazing if enough people stick to it for long enough that even one of its coinbases matures and becomes spendable.  That will take at least four months, but is likely not to happen at all unless there are a whole lot of people with a whole lot of hashing power who don't care about making coins on the main branch of the fork.   I am tempted to maintain a pre-fork full node just out of morbid curiosity, but it would be madness to devote resources to mining it.

Cryddit

[Buffer Overflow]

The question you have to ask yourself is, will you ever actually get sidechains if you stay on the 1MB chain?  The anti-fork crowd have gone on record as saying they'd like to set up a bunch of nodes running the current version that can be left running and basically forgotten about, so if there are new features people want to see added, you'll be out of luck if the network isn't updated to support that new feature.  

No.  You won't get side-chains if you stay on the 1MB block chain, because the 1MB block chain will not move fast enough to support them.

B.. Bu.. But.. But... Anti-fork brigade told me side-chains would solve everything. I feel so lied to.

[AMVM]

(...) You won't get side-chains if you stay on the 1MB block chain, because the 1MB block chain will not move fast enough to support them. (...)

Can you elaborate? First person I hear making this claim.

[Cryddit]

(...) You won't get side-chains if you stay on the 1MB block chain, because the 1MB block chain will not move fast enough to support them. (...)

Can you elaborate? First person I hear making this claim.

Even after a new version of the software containing a hard fork is deployed, the new protocol does not actually go into effect (the hard fork does not actually happen) until more than 95% of the hashing power is ready to go onto the new protocol.  

That leaves less than 5% of the hashing power on the old protocol, meaning a new block on the old protocol could be found less than once every 3 hours.  Because that's clearly a loss and a failure, at least 90% of the remaining people would abandon it immediately, leaving it getting one block about every 33 hours.  Making it even more a loss and a failure, but assuming that last half-percent of the hashing power can be an irrational lunatic fringe who never ever leave, then with that amount of constant hashing power....  

The transaction rate on the old chain is one transaction per 86 seconds.  

Most transactions never get into a block on the old chain, because the blocks are only 1MByte and only happen once per 33 hours.  

If a transaction does happen to get into a block on the old chain, it will be ten days before it gets to confirmation depth.

A coinbase from finding a block on the old chain would take four and a half months to become spendable.

And 2016 blocks, which is the time between difficulty adjustments, flies by in just seven years and eight months.  The maximum adjustment an unforked bitcoin will allow is for the block speed to quadruple.  Therefore the time to each new difficulty adjustment is a quarter the time needed for the previous one.  Again assuming hashing power stays constant, which it won't:

You spend seven and two-thirds years getting blocks once per 33 hours.      (1 tx per ~86 seconds)
You spend a year and nine-tenths getting blocks once every 8 1/3 hours.        (1 tx per ~21 seconds)
You spend five months and 20 days getting blocks once every 2 1/12 hours.   (1 tx per ~5 seconds)
You spend a month and twelve days getting blocks once every half-hour.         (1 tx per second)

And after that difficulty adjustment works "normally".  It takes ten years to get back to handling 1 tx per second.

[zvs]

I see a fair amount of 900KB+ blocks now, so, yeah, the max needs to increase.  Dunno about how 20MB was chosen, though.  Why not 5MB and fork it again in a couple years?

Your 'average home user' likely hasn't been running a bitcoin node for a year or longer.   I mean, damn, one person downloading blockchain from me = saturated (768kbps).  Not only that, you wouldn't want to be stuck downloading the blockchain from me, because it'd be slow as hell.  It's better for me not to run a node, or at least not listen to incoming connections.

imo, a portion of transaction fees should be distributed to full nodes based on some criteria.  though unsure what that criteria would be to stop people from exploiting it.    but i digress

yes, block size needs to be increased

[DeathAndTaxes]

Cryddit's scenario also only considers a neutral scenario where the two networks go their separate ways.  Users who have coins prior to the fork have the same coins on both sides of the fork as well.  If one believes that the minority fork will die and thus the coins on that fork have no value they could be used to pay txn fees to buy up all the block space and help that inevitability occur much quicker.

Pre-fork coins however are valid on both networks which would mean making worthless transactions would result in losing coins on both forks due to fees.  One would obviously want to ensure these high fee transactions are valid only on the minority network.  The first thing would be to get a single output which is only valid on one chain.  Either the majority or the minority works so lets just look at the minority network.  The newly mined coins after the fork are only valid that chain.  For example at say block height 400,000 there will be two blocks 400,000a and 400,000b.  The coins from the coinbase reward in 400,000a can't be spent on the b network because they don't exist.  Even a single satoshi of post-fork coins can be used to turn pre-fork coins into post-fork coins.  A transaction which has as its inputs any number of pre-fork inputs plus at least one post fork input will create outputs which are spendable only on one of the forks.

100 BTC of prefork coins + 1 satoshi of post fork minority coinbase coins = 100.00000001 BTC worth of minority only coins.

Since this transaction can't exist on the majority network any txns which occur 'downstream' from it also can't exist on the majority network.  You could now spend this 100 BTC in fees on the minority network and still have the 100 BTC on the majority network.

Cryddit assumes after the fork occurs some miners will abandon it but lets be generous and assume that the full 5% continue to use the original network (if more than 5% do not support the fork it won't happen).  Lets also be generous and assume all build nothing but 1MB blocks (something miners today don't even do).  That would be only 7.2MB per day in block space.  To create paying txns which use up the entire block would only be 0.72 BTC per day in fees.  So just keep creating txns sending coins back to yourself paying a fee and use ensure all blocks are full.  Legitimate users however can pay higher than the minimum fee so it is just a question of how high you want to push the fees.   If you make txns with 10x the normal fee then everyone else will need to pay more than 10x the normal fee to get confirmations and that would only cost 7.2 BTC per day in fees.  

In the "1MB is a great idea" post I pointed out that if txn demand exceeds block capacity by some significant margin it will occur off-chain on centralized systems and those centralized systems will eventually consume all the block space for settlement transactions.  Since they can outpay direct users in txns fees they will squeeze out the direct users.  The high fee txns is a good way to illustrate that at no cost on the minority fork.  I think someone in this thread said Bitcoin would be great even if it cost $5 per transaction.  If the average txn is 500 bytes then $5 per txn would be $10,000 per MB or 288 BTC spent in fees per day.  That might be much for a single user to do but a group of users working together could do it.  No legitimate txn would ever be confirmed no matter how long they waited unless they paid more.

DISCLAIMER: This is more an academic exercise than a practical suggestion.  Despite all the FUD I think when the fork happens the other chain will simply die off.  I would strongly discourage performing this unless you completely understand the concept of transaction validity across parallel forks.  If you think Bitcoin works on addresses and balances you will probably lose your Bitcoins on both forks by creating txns which are valid on both forks.

[NewLiberty]

(...) You won't get side-chains if you stay on the 1MB block chain, because the 1MB block chain will not move fast enough to support them. (...)

Can you elaborate? First person I hear making this claim.

Even after a new version of the software containing a hard fork is deployed, the new protocol does not actually go into effect (the hard fork does not actually happen) until more than 95% of the hashing power is ready to go onto the new protocol.  

That leaves less than 5% of the hashing power on the old protocol, meaning a new block on the old protocol could be found less than once every 3 hours.  Because that's clearly a loss and a failure, at least 90% of the remaining people would abandon it immediately, leaving it getting one block about every 33 hours.  Making it even more a loss and a failure, but assuming that last half-percent of the hashing power can be an irrational lunatic fringe who never ever leave, then with that amount of constant hashing power....  

The transaction rate on the old chain is one transaction per 86 seconds.  

Most transactions never get into a block on the old chain, because the blocks are only 1MByte and only happen once per 33 hours.  

If a transaction does happen to get into a block on the old chain, it will be ten days before it gets to confirmation depth.

A coinbase from finding a block on the old chain would take four and a half months to become spendable.

And 2016 blocks, which is the time between difficulty adjustments, flies by in just seven years and eight months.  The maximum adjustment an unforked bitcoin will allow is for the block speed to quadruple.  Therefore the time to each new difficulty adjustment is a quarter the time needed for the previous one.  Again assuming hashing power stays constant, which it won't:

You spend seven and two-thirds years getting blocks once per 33 hours.      (1 tx per ~86 seconds)
You spend a year and nine-tenths getting blocks once every 8 1/3 hours.        (1 tx per ~21 seconds)
You spend five months and 20 days getting blocks once every 2 1/12 hours.   (1 tx per ~5 seconds)
You spend a month and twelve days getting blocks once every half-hour.         (1 tx per second)

And after that difficulty adjustment works "normally".  It takes ten years to get back to handling 1 tx per second.

If you assume your conclusion, why bother with the proof?

[wilth1]

VISA and religion aside, difficulty optimizes mining for an ideal market condition. What desirable mining environment should block markets maximize?

A)Potential for inclusion of any and all transactions always
B)Immediate space for 40x current volume with a blind increase and another eventual hard limit
C)Inclusion of most non-spam* transactions most of the time, based on necessity
D)A scarcity magnifying limit that caps transaction volume and velocity starting soon

Most users don't want D. Non-generating nodes don't want A or probably B. Generating nodes want the most profitable [valuable] thing possible. Adversaries want A or D

*Most would agree that creating volume to intentionally fill blocks and slow the network qualifies as "spam"

[zvs]

20MB seems excessive, but which pools would actually use 20MB?  It seems like most are still stuck on 750KB (default) instead of increasing it to 1MB.

If by some miracle a backlog of 20MB of transactions occurred, any pool that spit out a 20MB block, well, I wouldn't be surprised if the orphan risk outweighed any gains from transaction fees.

....  I still think it should be a more gradual increase, but since mining is, so, well, centralized, there is only some half a dozen or maybe a dozen people tops that matter in terms of setting actual sizes & transaction fees.

[danielpbarron]

Pre-fork coins however are valid on both networks which would mean making worthless transactions would result in losing coins on both forks due to fees

False. If you have pre-fork coins, you have equal amounts of post-fork coins on each chain. There is no "worthless transaction." You can safely split the coins and spend them both. The split is achieved by broadcasting a transaction on one chain and a doublespend of it on the other. If you don't doublespend it, someone else will probably leak the transaction to the other chain; you will still have coins on both, and can just try to split again until it works. There is no risk of accidentally making your coins unspendable, and I suspect you have implied this possibility for the purpose of tricking people into not trying. I know it's tempting to believe that your stupid fork will happen because of the overwhelming amount of USG clones spamming this thread, but the fact of the matter is this: #bitcoin-assets will win the fork war, and the doublespending of coins will only benefit us.

[traincarswreck]

False. If you have pre-fork coins, you have equal amounts of post-fork coins on each chain. There is no "worthless transaction." You can safely split the coins and spend them both. The split is achieved by broadcasting a transaction on one chain and a doublespend of it on the other. If you don't doublespend it, someone else will probably leak the transaction to the other chain; you will still have coins on both, and can just try to split again until it works. There is no risk of accidentally making your coins unspendable, and I suspect you have implied this possibility for the purpose of tricking people into not trying. I know it's tempting to believe that your stupid fork will happen because of the overwhelming amount of USG clones spamming this thread, but the fact of the matter is this: #bitcoin-assets will win the fork war, and the doublespending of coins will only benefit us.

20 years a man gave lectures on exactly how bitocin will do exactly this to our civilization, and we would rather ignore logic and reason.  People casting aside netwon, copernicus, Nash, as if they know this subject at all.

But the famous classical "Gresham’s Law" also reveals the intrinsic difficulty. Thus "good money" will not naturally supplant and replace "bad money" by a simple Darwinian superiority of competitive species. Rather than that, it must be that the good things are established by the voluntary choice of human agencies. And these resp-onsible agencies, being naturally of the domain of polit-ically derived authorities, would need to make appropriate efforts to achieve such a goal and to pay the costs that are entailed before their societies can benefit. And the benefits would come from the improvement in the quality
of this public utility (money) which serves to facilitate the game-theoretic function of "the transfer of utility".

What is so broken that we feel so privileged to act so ignorant and blind? Should we function on science or religion?

[danielpbarron]

because of the overwhelming amount of USG clones spamming this thread

20 years a man gave lectures on exactly how bitocin will do exactly this to our civilization, and we would rather ignore logic and reason.  People casting aside netwon, copernicus, Nash, as if they know this subject at all.

What is so broken that we feel so privileged to act so ignorant and blind? Should we function on science or religion?

Here's a good example of what I mean. Words strung together with the resemblance of a valid point, but on closer inspection there is no appropriate context. This is one of many USG spam accounts put here to create the illusion of a debate.

[traincarswreck]

Here's a good example of what I mean. Words strung together with the resemblance of a valid point, but on closer inspection there is no appropriate context. This is one of many USG spam accounts put here to create the illusion of a debate.

Firstly this is to be a dialogue not a debate (to which I have extensively defined especially in relation to what is ideal).  2nd I am by far the most knowledgeable one here on the interrelation of these subjects especially in regards to "what is ideal money".  If you have been following the dialog you will also notice I have produced the most content, and especially at least up to the point that it looks absolutely trollish for you to suggest that someone created me as a "spam account" (ie I have proof of work).

It might be others don't appreciate my view and my content, but I have functioned from pure logical and fact.  Produced related content that nearly no one, or no one here has been previously made aware of...

I for one would appreciate, that you don't...and that we resume what is called "dialogue", and begin to discuss what is logical and based on fact.

Cliffs: my content is organized on a blog, but it is clearly too big for some.

[traincarswreck]

We should press on:

In the beginning, people won’t trust each other. But I think that if they see the importance of the dialogue, they will work with it.

The object of a dialogue is not to analyse things, or to win an argument, of to exchange opinions. Rather, it is to suspend your opinions and to look at the opinions-to listen to everybody’s opinions, to suspend them, and to see what all that means.

The point is not to establish a fixed dialogue group forever, but rather one that lasts long enough to make a change. If you keep holding it for too long, it may become caught up in habits again. But you have to keep it up for awhile, or else it won’t work

“The voters in the U.K. are expecting to have the opportunity to vote in a referendum relating to the adoption, for the U.K., of the euro (which is already adopted in Ireland). Here they have a dramatic conflict, since the pound was the original currency of “the gold standard”, with its value pegged to gold in 1717 by Isaac Newton who was then Master of the Mint. (Of course it was not irrelevant that George II, the king then, was an early Hanoverian and also ruled territory in Germany.)”

Who then, can tell us, why this point about George II being an early Hanoverian, is "of course" "not irrelevant"?

That is seemingly where we should go with this next...

[zvs]

We should press on:

In the beginning, people won’t trust each other. But I think that if they see the importance of the dialogue, they will work with it.

The object of a dialogue is not to analyse things, or to win an argument, of to exchange opinions. Rather, it is to suspend your opinions and to look at the opinions-to listen to everybody’s opinions, to suspend them, and to see what all that means.

The point is not to establish a fixed dialogue group forever, but rather one that lasts long enough to make a change. If you keep holding it for too long, it may become caught up in habits again. But you have to keep it up for awhile, or else it won’t work

“The voters in the U.K. are expecting to have the opportunity to vote in a referendum relating to the adoption, for the U.K., of the euro (which is already adopted in Ireland). Here they have a dramatic conflict, since the pound was the original currency of “the gold standard”, with its value pegged to gold in 1717 by Isaac Newton who was then Master of the Mint. (Of course it was not irrelevant that George II, the king then, was an early Hanoverian and also ruled territory in Germany.)”

Who then, can tell us, why this point about George II being an early Hanoverian, is "of course" "not irrelevant"?

That is seemingly where we should go with this next...

Is this one of those $0.02USD HITs on Mechanical Turk?

[Cryddit]

No, not quite a mechanical-turkism although it might as well be as far as the present discussion is concerned.

I sort of get it, but Traincarswreck is thinking at right angles to what the rest of us consider relevant, and writing for an audience of lurkers, monitors, eavesdroppers, or historians who are or will be reading this thread for other purposes, rather than writing to be read and responded to by any of us who are actually participating in realtime.  

As to whether the communication will be or is considered meaningful by any such audience, I'm confident that 95% of them will have roughly the same opinion of Traincarswreck's relevance as 99.8% of us.  Traincarswreck is hoping that the 5% of unseen listeners with the greatest insight into his (or her, or their) posts will also be the most influential in their respective spheres.  Which I'd consider to be unwarranted optimism on his (or her, or their) part, but he's entitled to hope.

I could try to explain further, but there's at least a 75% chance I'd be wrong about the objectives, intended audiences, and means.  Besides I don't think I care enough.  

Cryddit

Note:  All statistics and percentages given in this post are estimates and guesses which exist in the absence of measurement and proof.  In other words, I'm just guessing.