Anti-fork guys: What is YOUR proposal?

[manselr]

This is the situation and what we know:



So we are heading towards a dead end. Now, Gavin has proposed a fork of 20MB. That's a proposal. I don't know for how long would 20MB last for. Ideally, it should be able to deal with all the world's biggest electronic payment platforms transactions convined, so, can 20MB deal with a volume of for example, Western Union, Paypal, Mastercard... and so on and so on, combined? Because that is were we should aim at for the future. Bitcoin must be able to engulf all of that volume within itself to become the #1 electronic transaction method on the planet. Everything else is bullshit.
Will 20MB be enough for this or we will need ANOTHER fork in the future? If 20MB is enough, it's a reasonable proposal. Sure, forking sucks, but im not hearing any other realistic solutions. Of course, like I said before, im assuming 20MB is the definitive fork. We can't be forking every X years, that's just retarded.

What is not reasonable, is making a ton of FUD threads, saying how Gavincoin will kill BTC, while implying staying with the 1MB blocksize limit is all fine and dandy. So everytime you say Gavin's proposal is shit, explain why AND propose your own solution, otherwise you are just spreading cheap FUD.

Another question I have: What is Satoshi's view on this? im sure he expected this to happen, so I guess he had an opinion before leaving.

[1714830207]

1714830207

[1714830207]

1714830207

[1714830207]

1714830207

[1714830207]

1714830207

[1714830207]

1714830207

[gentlemand]

20mb blocks would be a microscopic slice of global payment volume. 'Kicking the can down the road' is the phrase that pops up regularly when discussing it.

Hopefully if there is a fork it'll be one to last for a long, long time. There's talk of an ever increasing block size year on year to be implemented too. I'll leave it to the experts to fight it out amongst themselves.

[NyeFe]

"Hold your heads under the table and hope we never require a block size increase, yes as adaptation grows"

SMH 

[oblivi]

I dont think we'll see many replies here, people are too busing making fud threads while not proposing any solution. Then again, what gentlemand said is a problem, if we'll need more and more forks it doesnt sound good. Hopefully if they make a fork, it's a fork that is self adaptive to this problem so we don't need to ever fork again.

[Fuserleer]

I'm not anti-fork, nor am I really bothered which "stick" everyone ends up running with, but the problem remains that BTC can not scale to be a general world currency.

The only solution is a distributed ledger of some form, but on a single block chain, that is not possible to do.  Side chains could be classified as a form of distributed ledger, and may work to some degree of success, but its more of form of partitioning rather than distribution, because you can not be sure that all side chains are using compatible properties or protocols to directly interact with each other, and so there is additional overhead to enforce this.

To achieve load capabilities that will allow any crypto-currency to truly become a mass market, general world currency, and stay true to the original ideals of Bitcoin as a (100% decentralized), said currency must be developed from the beginning with a viable, working distributed ledger solution in mind.

With BTC, there is no solution, period.  Increasing the block size will never solve this problem if BTC was to become truly mainstream, due to the required block size to cope with 100's - 1000's of transactions per second loading.  If block sizes are in the 100MB+ to deal with that kind of load, the end result is more centralization, and less full nodes on the network due to the costs and inconvenience of maintaining/running a full node.  The block chain size can be mitigated with some form of pruning, but you still have the bandwidth and processing requirements to ensure full nodes are able to deal with those high transaction loads.

[not altcoin hitler]

I'm anti fork an my proposal is to just fire Gavin and be happy thereafter.
And he should take his fanclub with him!

[Amph]

I dont think we'll see many replies here, people are too busing making fud threads while not proposing any solution. Then again, what gentlemand said is a problem, if we'll need more and more forks it doesnt sound good. Hopefully if they make a fork, it's a fork that is self adaptive to this problem so we don't need to ever fork again.

the proposed alternative are not good enough, hence the only solution is to go ahead and making this change, other 4 devs are worried that in the future it might be necessary to fork again because 20mb would not be enough, but with 20MB we are in the same boat as visa or mastercard(if i'm not mistaken, don't remember they number of transaction per second), talking about transaction per second, worse case we should do 30 mb right now

[LiteCoinGuy]

I dont think we'll see many replies here, people are too busing making fud threads while not proposing any solution. Then again, what gentlemand said is a problem, if we'll need more and more forks it doesnt sound good. Hopefully if they make a fork, it's a fork that is self adaptive to this problem so we don't need to ever fork again.

the proposed alternative are not good enough, hence the only solution is to go ahead and making this change, other 4 devs are worried that in the future it might be necessary to fork again because 20mb would not be enough, but with 20MB we are in the same boat as visa or mastercard(if i'm not mistaken, don't remember they number of transaction per second), talking about transaction per second, worse case we should do 30 mb right now

that is the reason why we should make an automatic increase. dont forget, we are winning time (years) and other solutions will be developed too!

[Fuserleer]

I dont think we'll see many replies here, people are too busing making fud threads while not proposing any solution. Then again, what gentlemand said is a problem, if we'll need more and more forks it doesnt sound good. Hopefully if they make a fork, it's a fork that is self adaptive to this problem so we don't need to ever fork again.

the proposed alternative are not good enough, hence the only solution is to go ahead and making this change, other 4 devs are worried that in the future it might be necessary to fork again because 20mb would not be enough, but with 20MB we are in the same boat as visa or mastercard(if i'm not mistaken, don't remember they number of transaction per second), talking about transaction per second, worse case we should do 30 mb right now

20MB gets Bitcoin nowhere even close to VISA/Mastercard processing abilities.

Visa on a regular day processes 2-3000 tx/s, and has a peak capacity of about 40,000 tx/s

20MB blocks give Bitcoin a theoretical capacity of about 140 tx/s, in practice it will be more around the 70tx/s

For Bitcoin to have the same capacity as VISA alone on a normal day, would require block sizes of around 500MB, to be able to handle VISA peak of 40,000 tx/s, blocks would need to be about 10GB

[gentlemand]

the proposed alternative are not good enough, hence the only solution is to go ahead and making this change, other 4 devs are worried that in the future it might be necessary to fork again because 20mb would not be enough, but with 20MB we are in the same boat as visa or mastercard(if i'm not mistaken, don't remember they number of transaction per second), talking about transaction per second, worse case we should do 30 mb right now

20mb is more but it's still not anywhere near the same ball park. If we assume it's 20 times larger than the theoretical 7 transaction per second, but it appears to be 2.7 really, then that's 280 tps in a best case scenario.

Visa can handle 24,000 according to themselves.

I don't think they're directly comparable. If Bitcoin really was an everyday currency for a reasonable percentage of the world it would be knocking up far larger numbers.

[pedrog]

Blocksize, as is today, already cannot deal with all bitcoin transactions, you forgot many transactions already happen off chain, in Coinbase, Xapo, Bitpay, etc, not only within their service but between services that already have an agreement.

We don't need one solution, we need several solutions to scale Bitcoin to compete with Paypal, Visa and Mastercards of the world.

We can have 20 MB blocks and we can have the Lightning Network and we can have off chain transactions, because we need all of those if we're gonna scale to the moon.

[Amph]

I dont think we'll see many replies here, people are too busing making fud threads while not proposing any solution. Then again, what gentlemand said is a problem, if we'll need more and more forks it doesnt sound good. Hopefully if they make a fork, it's a fork that is self adaptive to this problem so we don't need to ever fork again.

the proposed alternative are not good enough, hence the only solution is to go ahead and making this change, other 4 devs are worried that in the future it might be necessary to fork again because 20mb would not be enough, but with 20MB we are in the same boat as visa or mastercard(if i'm not mistaken, don't remember they number of transaction per second), talking about transaction per second, worse case we should do 30 mb right now

20MB gets Bitcoin nowhere even close to VISA/Mastercard processing abilities.

Visa on a regular day processes 2-3000 tx/s, and has a peak capacity of about 40,000 tx/s

20MB blocks give Bitcoin a theoretical capacity of about 140 tx/s, in practice it will be more around the 70tx/s

For Bitcoin to have the same capacity as VISA alone on a normal day, would require block sizes of around 500MB, to be able to handle VISA peak of 40,000 tx/s, blocks would need to be about 10GB

tnx for giving me the exact numbers, then if those are correct we only need another fork for 400-500, with another rise of x20, i don't find this to be a big problem like the others 4 dev are claiming, and complaining about

[pereira4]

I'm not anti-fork, nor am I really bothered which "stick" everyone ends up running with, but the problem remains that BTC can not scale to be a general world currency.

The only solution is a distributed ledger of some form, but on a single block chain, that is not possible to do.  Side chains could be classified as a form of distributed ledger, and may work to some degree of success, but its more of form of partitioning rather than distribution, because you can not be sure that all side chains are using compatible properties or protocols to directly interact with each other, and so there is additional overhead to enforce this.

To achieve load capabilities that will allow any crypto-currency to truly become a mass market, general world currency, and stay true to the original ideals of Bitcoin as a (100% decentralized), said currency must be developed from the beginning with a viable, working distributed ledger solution in mind.

With BTC, there is no solution, period.  Increasing the block size will never solve this problem if BTC was to become truly mainstream, due to the required block size to cope with 100's - 1000's of transactions per second loading.  If block sizes are in the 100MB+ to deal with that kind of load, the end result is more centralization, and less full nodes on the network due to the costs and inconvenience of maintaining/running a full node.  The block chain size can be mitigated with some form of pruning, but you still have the bandwidth and processing requirements to ensure full nodes are able to deal with those high transaction loads.

But didn't satoshi wanted Bitcoin to (ideally) become a world currency capable of dealing with said volume? I don't think he wanted Bitcoin to stay some small niche payment method.

[gentlemand]

But didn't satoshi wanted Bitcoin to (ideally) become a world currency capable of dealing with said volume? I don't think he wanted Bitcoin to stay some small niche payment method.

Here's his statement about it - "The current system where every user is a network node is not the intended configuration for large scale. That would be like every Usenet user runs their own NNTP server. The more burden it is to run a node, the fewer nodes there will be. Those few nodes will be big server farms"

Dunno whether he actually approved of that idea or whether it was an observation of future fact.

[jaberwock]

I dont think we'll see many replies here, people are too busing making fud threads while not proposing any solution. Then again, what gentlemand said is a problem, if we'll need more and more forks it doesnt sound good. Hopefully if they make a fork, it's a fork that is self adaptive to this problem so we don't need to ever fork again.

the proposed alternative are not good enough, hence the only solution is to go ahead and making this change, other 4 devs are worried that in the future it might be necessary to fork again because 20mb would not be enough, but with 20MB we are in the same boat as visa or mastercard(if i'm not mistaken, don't remember they number of transaction per second), talking about transaction per second, worse case we should do 30 mb right now

that is the reason why we should make an automatic increase. dont forget, we are winning time (years) and other solutions will be developed too!

how deal with people that "forget" to upgrade their clients or can only do it at a high cost?

Serious financial software can't be upgraded every day

[Fuserleer]

I dont think we'll see many replies here, people are too busing making fud threads while not proposing any solution. Then again, what gentlemand said is a problem, if we'll need more and more forks it doesnt sound good. Hopefully if they make a fork, it's a fork that is self adaptive to this problem so we don't need to ever fork again.

the proposed alternative are not good enough, hence the only solution is to go ahead and making this change, other 4 devs are worried that in the future it might be necessary to fork again because 20mb would not be enough, but with 20MB we are in the same boat as visa or mastercard(if i'm not mistaken, don't remember they number of transaction per second), talking about transaction per second, worse case we should do 30 mb right now

20MB gets Bitcoin nowhere even close to VISA/Mastercard processing abilities.

Visa on a regular day processes 2-3000 tx/s, and has a peak capacity of about 40,000 tx/s

20MB blocks give Bitcoin a theoretical capacity of about 140 tx/s, in practice it will be more around the 70tx/s

For Bitcoin to have the same capacity as VISA alone on a normal day, would require block sizes of around 500MB, to be able to handle VISA peak of 40,000 tx/s, blocks would need to be about 10GB

tnx for giving me the exact numbers, then if those are correct we only need another fork for 400-500, with another rise of x20, i don't find this to be a big problem like the others 4 dev are claiming, and complaining about

No problem.

It could become a big problem, read my post further up in this thread.  If Bitcoin was to become truly mainstream in the next 1-2 years, then technologies required such as connection bandwidth, CPU power, and storage to support that load, could be priced out of the "guy on the street" budget range.  If normal users are not able, or willing, to run full nodes and ensure real decentralization, then you end up in a mess.

I'm not anti-fork, nor am I really bothered which "stick" everyone ends up running with, but the problem remains that BTC can not scale to be a general world currency.

The only solution is a distributed ledger of some form, but on a single block chain, that is not possible to do.  Side chains could be classified as a form of distributed ledger, and may work to some degree of success, but its more of form of partitioning rather than distribution, because you can not be sure that all side chains are using compatible properties or protocols to directly interact with each other, and so there is additional overhead to enforce this.

To achieve load capabilities that will allow any crypto-currency to truly become a mass market, general world currency, and stay true to the original ideals of Bitcoin as a (100% decentralized), said currency must be developed from the beginning with a viable, working distributed ledger solution in mind.

With BTC, there is no solution, period.  Increasing the block size will never solve this problem if BTC was to become truly mainstream, due to the required block size to cope with 100's - 1000's of transactions per second loading.  If block sizes are in the 100MB+ to deal with that kind of load, the end result is more centralization, and less full nodes on the network due to the costs and inconvenience of maintaining/running a full node.  The block chain size can be mitigated with some form of pruning, but you still have the bandwidth and processing requirements to ensure full nodes are able to deal with those high transaction loads.

But didn't satoshi wanted Bitcoin to (ideally) become a world currency capable of dealing with said volume? I don't think he wanted Bitcoin to stay some small niche payment method.

I can't speak for Satoshi, nor his long term intentions, but you need to remember that Bitcoin was/is just experiment.  

If I was Satoshi and had all these ideas, Bitcoin in its current incarnation would of been the first step, a proof of concept, and not the final end game.  I'm certain that he was aware of its limitations, he was a smart guy, so would also know that to achieve global currency status would require some later refinement, revisions and work.  

I think what is more likely is that Bitcoin ran away from him before he had a chance to get to that point, it got more attention than he thought it would and got nervous about the implications.

[SpanishSoldier]

Proposal 1: Show some respect to Satoshi. Make it 21MB at least.

Proposal 2: I need to understand the lightning.network before proposing this part.

[pedrog]

I dont think we'll see many replies here, people are too busing making fud threads while not proposing any solution. Then again, what gentlemand said is a problem, if we'll need more and more forks it doesnt sound good. Hopefully if they make a fork, it's a fork that is self adaptive to this problem so we don't need to ever fork again.

the proposed alternative are not good enough, hence the only solution is to go ahead and making this change, other 4 devs are worried that in the future it might be necessary to fork again because 20mb would not be enough, but with 20MB we are in the same boat as visa or mastercard(if i'm not mistaken, don't remember they number of transaction per second), talking about transaction per second, worse case we should do 30 mb right now

that is the reason why we should make an automatic increase. dont forget, we are winning time (years) and other solutions will be developed too!

how deal with people that "forget" to upgrade their clients or can only do it at a high cost?

Serious financial software can't be upgraded every day

That's why Gavin already began this discussion, because it will take many months for the majority of the network to upgrade, that's the thing with distributed networks, one has to plan it way ahead.

[Benjig]

I'm anti fork an my proposal is to just fire Gavin and be happy thereafter.
And he should take his fanclub with him!

Ok and what will you do when world demands more Bitcoin transactions, if we want to make Bitcoin bigger than the baby already it is we need changes. Please do not be afraid of change, that will kill you..

More forks in the future may be needed but I'm not sure if FIAT will dissapear leaving economics world to Bitcoin or any other...

[celestio]

I'm not anti-fork, nor am I really bothered which "stick" everyone ends up running with, but the problem remains that BTC can not scale to be a general world currency.

The only solution is a distributed ledger of some form, but on a single block chain, that is not possible to do.  Side chains could be classified as a form of distributed ledger, and may work to some degree of success, but its more of form of partitioning rather than distribution, because you can not be sure that all side chains are using compatible properties or protocols to directly interact with each other, and so there is additional overhead to enforce this.

To achieve load capabilities that will allow any crypto-currency to truly become a mass market, general world currency, and stay true to the original ideals of Bitcoin as a (100% decentralized), said currency must be developed from the beginning with a viable, working distributed ledger solution in mind.

With BTC, there is no solution, period.  Increasing the block size will never solve this problem if BTC was to become truly mainstream, due to the required block size to cope with 100's - 1000's of transactions per second loading.  If block sizes are in the 100MB+ to deal with that kind of load, the end result is more centralization, and less full nodes on the network due to the costs and inconvenience of maintaining/running a full node.  The block chain size can be mitigated with some form of pruning, but you still have the bandwidth and processing requirements to ensure full nodes are able to deal with those high transaction loads.

I've been looking at both sides of the equation and I have to agree. The proposal to just raise the blocksize limit to 20mb isn't solving anything, it's just prolonging the issue(A careless quick temporary fix if you will). Bitcoin still won't be able to scale.