Hards forks are the real issue, not blockchain size

[Cubic Earth]

Blockchain size is a real and current concern of course, and in the future there will be many more hard-coded issues that will arise. Hard forks happen and will continue to happen, as we see in Solidcoin and all the rest.  The developers should no be in any position to bring down bitcoin, because if they are, bitcoin will eventually fail.  Bitcoin is about faith in the market system.  The system must be resistant to bad ideas.  Is it?

Someone code up a hard-fork and see if it takes hold.  That is the bitcoin way and it is not too be feared.

[Come-from-Beyond]

There. Is. Another. Way.

[hazek]

I actually don't disagree with the OP. Except that all this drama around the block size limit is nothing but a market regulated strictly by consumption (i.e. a free market) actually regulating itself. All these threads, all the whining and squealing and protesting and lobbying, ect., all this is part of the process of how this market will through levels of consumption figure out which fork of Bitcoin best serves it - hint: it will most likely be the one that will be the most consumed. And hey, there's nothing that says that two forks can't coexist.. Yes, maybe it would damage Bitcoin a bit but perhaps BitcoinA and BitcoinB can both recover and be their own seperate thing and find equal demand in consumption.

I kind of already have a wish list for my ideal cryptocurrency that I will put into words and post in the alternative cryptocurrencies section in order to invite devs to take the Bitcoin source code and actually try and innovate and find some improvements which might yield a successful Bitcoin 2.0. So far I'm not impressed with any of the alternatives because none of them solve anything groundbreaking.

[JeromeS]

I agree. I've been thinking about this, here's what I'm at right now,

As long as the block chain *does* fork, everyone's coin also forks with it, right ? so people can just sell their coin on one chain to buy more on the other if they want to. Everyone is satisfied.


Two problems I can think of though,

1- what if the fork doesn't happen for some reason ? I don't know what would prevent it from happening but people on other threads are saying it needs "miner consensus". I would appreciate it if someone can explain this to me.

2- This could be a silly question, but is there something to prevent someone from taking transactions from one chain and rebroadcasting them on the other ?

[DannyHamilton]

1- what if the fork doesn't happen for some reason ? I don't know what would prevent it from happening but people on other threads are saying it needs "miner consensus". I would appreciate it if someone can explain this to me.

You need consensus if you want to implement a "forking" change without actually causing the blockchain to fork into two incompatible systems. If you actually WANT to cause the blockchain to fork into two incompatible systems, you only need a single miner willing to mine the forked blockchain (although the more distributed the mining hash power is, the more secure your minority fork will be)

2- This could be a silly question, but is there something to prevent someone from taking transactions from one chain and rebroadcasting them on the other ?

To transmit a transaction, you need specific outputs received at an address to spend.  This means that bitcoins that were received prior to the fork will be spendable on both forks, while bitcoins that are received after the fork will only be spendable on the fork where they are received.

[Cubic Earth]

You need consensus if you want to implement a "forking" change without actually causing the blockchain to fork into two incompatible systems. If you actually WANT to cause the blockchain to fork into two incompatible systems, you only need a single miner willing to mine the forked blockchain (although the more distributed the mining hash power is, the more secure your minority fork will be)

This is just a myth, or possibly circular logic, that has been repeated many times on these forums.  To fork, You actually just need one computer, or possibly two, running something that is hard-coded differently than is the current bitcoin that we are all using currently and that builds of the blockchain that we all know and love.  The question is, how many people would make the switch to the other system.  I would guess the current bitcoin will last from 6 months to 1 year, and then the vast majority of users will switch to something else.  Would we probably start to define "Bitcoin" at that point as what ever system the masses switched to.  But that could really be anything.  There is no guarantee or promise with the current system, as that very concept would suggest there is some entity which made a commitment.  Perhaps the only promise is that such an entity does not exist. 

And hey, there's nothing that says that two forks can't coexist.

And two forks already exist!  It's just the original on is still going strong.

...hint: it will most likely be the one that will be the most consumed.

Yes, and that is the only truth here.  If Mt. Gox decides to code it's own version, based on the blockchain to date, but going forward gives only Mt. Gox the ability to manage the integrity of the network, well if people decided they wanted to be part of that network instead of our current one, so be it.  Nothing would have been violated.

[DannyHamilton]

You need consensus if you want to implement a "forking" change without actually causing the blockchain to fork into two incompatible systems. If you actually WANT to cause the blockchain to fork into two incompatible systems, you only need a single miner willing to mine the forked blockchain (although the more distributed the mining hash power is, the more secure your minority fork will be)

This is just a myth, or possibly circular logic, that has been repeated many times on these forums.  To fork, You actually just need one computer, or possibly two, running something that is hard-coded differently than is the current bitcoin that we are all using currently and that builds of the blockchain that we all know and love.  The question is, how many people would make the switch to the other system.  I would guess the current bitcoin will last from 6 months to 1 year, and then the vast majority of users will switch to something else.  Would we probably start to define "Bitcoin" at that point as what ever system the masses switched to.  But that could really be anything.  There is no guarantee or promise with the current system, as that very concept would suggest there is some entity which made a commitment.  Perhaps the only promise is that such an entity does not exist. 

You aren't paying attention.  You just said the same thing as me while claiming that my statement is a myth?  Try being a bit more consistent.

[Cubic Earth]

You aren't paying attention.  You just said the same thing as me while claiming that my statement is a myth?  Try being a bit more consistent.

I think you expressed two points that contradicted themselves.  I agreed with one of your statements and disagreed with the other.  In a good-natured rebuttal, I would like to point out the irony in you suggesting that I be more consistent 

[DannyHamilton]

You aren't paying attention.  You just said the same thing as me while claiming that my statement is a myth?  Try being a bit more consistent.

I think you expressed two points that contradicted themselves.  I agreed with one of your statements and disagreed with the other.  In a good-natured rebuttal, I would like to point out the irony in you suggesting that I be more consistent 

I think we're just playing with semantics at this point, so I'll just stop here.

[JeromeS]

You need consensus if you want to implement a "forking" change without actually causing the blockchain to fork into two incompatible systems. If you actually WANT to cause the blockchain to fork into two incompatible systems, you only need a single miner willing to mine the forked blockchain (although the more distributed the mining hash power is, the more secure your minority fork will be)

So, as long as 2 people want for a 1MB fork to exist, there's nothing to prevent it ?

To transmit a transaction, you need specific outputs received at an address to spend.  This means that bitcoins that were received prior to the fork will be spendable on both forks, while bitcoins that are received after the fork will only be spendable on the fork where they are received.

I meant, say I have an output that exists on the pre-split part of both block chains, and I spend it on one network, sending it to someone else. What's to prevent that person from rebroadcasting that tx on the other network as well ?

[DannyHamilton]

You need consensus if you want to implement a "forking" change without actually causing the blockchain to fork into two incompatible systems. If you actually WANT to cause the blockchain to fork into two incompatible systems, you only need a single miner willing to mine the forked blockchain (although the more distributed the mining hash power is, the more secure your minority fork will be)

So, as long as 2 people want for a 1MB fork to exist, there's nothing to prevent it ?

Correct.

To transmit a transaction, you need specific outputs received at an address to spend.  This means that bitcoins that were received prior to the fork will be spendable on both forks, while bitcoins that are received after the fork will only be spendable on the fork where they are received.

I meant, say I have an output that exists on the pre-split part of both block chains, and I spend it on one network, sending it to someone else. What's to prevent that person from rebroadcasting that tx on the other network as well ?

Nothing.

[JeromeS]

Correct.

That's comforting...

Nothing.

Well, thanks for explaining.

[Monster Tent]

If a fork with 60% exists and one with 40% I would probably see bitcoin as a failed experiment.

[Cubic Earth]

To transmit a transaction, you need specific outputs received at an address to spend.  This means that bitcoins that were received prior to the fork will be spendable on both forks, while bitcoins that are received after the fork will only be spendable on the fork where they are received.

I meant, say I have an output that exists on the pre-split part of both block chains, and I spend it on one network, sending it to someone else. What's to prevent that person from rebroadcasting that tx on the other network as well ?

Nothing.

So if you had assets on each chain or the fork, residing at the same respective addresses, you would have to spend the simultaneously to avoid getting ripped off on the one of the chains?  I guess it would be a race down to the millisecond.  Someone would have software and a high speed connection tapping into both networks and attempting to take advantage of the smallest delay.

[Cubic Earth]

If a fork with 60% exists and one with 40% I would probably see bitcoin as a failed experiment.

Of course if the total number of users was 10x higher, each fork could have more people using it and more money flowing through it than the current system does.  I'm not sure that would be a failure.

[JeromeS]

So if you had assets on each chain or the fork, residing at the same respective addresses, you would have to spend the simultaneously to avoid getting ripped off on the one of the chains?  I guess it would be a race down to the millisecond.  Someone would have software and a high speed connection tapping into both networks and attempting to take advantage of the smallest delay.

I suppose if this is true there would still be a way around it. Send the coin to a different address that you own on each chain, wait for it to clear, wait a few blocks to make sure you're on the respective "main chains", then proceed as usual, all new transactions on each network would reference an output that doesn't exist on the other.

[JeromeS]

Of course if the total number of users was 10x higher, each fork could have more people using it and more money flowing through it than the current system does.  I'm not sure that would be a failure.

Agreed. Each network would only have to process and keep track of a fraction of the total transactions. This could actually be better for bitcoin.

It's fun to speculate about this. What if the networks start specializing ?

The 10Mb network would be less decentralized, slower, and more expensive to run with 10 times the needed bandwidth and disk space to keep track of hundreds of people betting their $1. Most blocks would have a lot of low value transactions with small or zero fees. The 1 Mb network on the other hand would handle all the serious money flow with less overhead.

[hazek]

If a fork with 60% exists and one with 40% I would probably see bitcoin as a failed experiment.

And how does that matter? How you'd see it I mean?

[jgarzik]

If a fork with 60% exists and one with 40% I would probably see bitcoin as a failed experiment.

Even at much smaller percentages, bitcoin's trust could evaporate.  Consider not knowing whether or not a particular bitcoin node would accept your transaction.  Or for merchants, the possibility exists of double-spending coins, one spend per fork.

Any long lived chain fork could have a hugely negative impact.

[Cubic Earth]

If a fork with 60% exists and one with 40% I would probably see bitcoin as a failed experiment.

Even at much smaller percentages, bitcoin's trust could evaporate.  Consider not knowing whether or not a particular bitcoin node would accept your transaction.  Or for merchants, the possibility exists of double-spending coins, one spend per fork.

Any long lived chain fork could have a hugely negative impact.

Why then, so you suppose the trust has already not evaporated?  There have already been forks.  One possible answer is "those forks have not been successful".  The answer would have to be, paradoxically, that only successful forks would somehow spell failure.  I don't think I would buy that argument, but please, explain you thinking further.