Solving the problem of on-chain scaling

[Andriian]

You can reduce a 1gb blockchain to less than 1kb.

You can, but on very specific scenario such as Zip bomb or the whole data have identical long sequence.

I have to say "False until proven real" on your claim

It looks like that there are projects utilizing Zero-Knoledge-Proof like Coda aiming to solve blockchain size problem (may be at some computation cost). Isn't it a promising approach?

Sounds interesting, but AFAIK & IMO (after quick research) it's alternative for light/SPV node (as compared with merkle tree, block header, bloom filter and neutrino).
Someone still need to store whole blockchain and send relevant information with ZK-snarks.

CMIIW.

It was my first impression that that they are trying to make new light/SPV approach but it seems they are not. If I got it right, 20 kB is what blockchain will look like in any validator, i.e. full node, and ZK magic will enable check any transaction against that 20 kB  (However, it looks like that essential computation power is needed to "pack" every new block via ZK).
Users need to store private keys (may be some small metadata about their balances as well) in their devices and the only entities which might need full blockchain are API providers and Blockchain explorers for historical/analytical data

[Khaos77]

Now explain why all the myriad altcoins which do exactly that are still a speck of dust in the wind compared to Bitcoin.  

Because of the price manipulators , which price is all your simple mind can conceive.

Who said anything about price?  I'm talking about utility.  A currency I'm happy to receive as a payment because I have the capability to purchase tangible goods and services with it.  Other coins can't match the network effects of Bitcoin, which is something you continually underestimate the importance of.  

As for your other remark about non-mining nodes and security, I'm not going to waste my time repeating something to you that I've said enough times in the past (to far superior opponents, no less).  Suffice to say, you are wrong.  End of.  

Did you go in a grocery store and pay with bitcoin, did you pay your electric bill with bitcoin,
did you pay for your cup of coffee with bitcoin.

Answer is NO, because even credit cards charge a lower fee than bitcoin current transaction fees.

Bitcoin is a buy and hold and sell (for the non-stupid.)
It has no useful utility as a payment service due to it fluctuations in price verses fiat and it fluctuations in transaction fees.
Also who had time to wait for 30 minutes for 3 confirmations at a gas station.   

* Your non-mining node serves no purpose except propping up your fragile ego, that you matter, you don't.  

[Andriian]

Did you go in a grocery store and pay with bitcoin, did you pay your electric bill with bitcoin,
did you pay for your cup of coffee with bitcoin.

Answer is NO, because even credit cards charge a lower fee than bitcoin current transaction fees.

Bitcoin is a buy and hold and sell (for the non-stupid.)
It has no useful utility as a payment service due to it fluctuations in price verses fiat and it fluctuations in transaction fees.
Also who had time to wait for 30 minutes for 3 confirmations at a gas station.   

* Your non-mining node serves no purpose except propping up your fragile ego, that you matter, you don't.  

There is another use case except buy/hold/save which is moving cash. In particular this is widely used by business to avoid regulator restrictions or slow international bank transfers which are often much more that 30 minutes.

[franky1]

those saying increasing the blocksize comes with a cost.. are outdated notions from years ago

its already been demythed about hard drive and internet costs. so no point trying to continue to try pulling that rabbit out of your hat... the trick has ben revealed

as for trying to compare bitcoin to altcoins that do allow more transaction count per day.. the myth iss ' but look people dont use them'.
1. this has nothing to do with the altcoins feature but because merchants are sticking with bitcoin because of market price
2. altcoins capable of large scale, tested the capacity and guess what, computers did not blow up or ask for a credit card number to upgrade its hardware
3. just the existance of such altcoins proves it can be done

bitcoin only uses under 200mb a day, yes a DAY (1.4mb * 144 blocks(most blocks are under 1.4mb))
just streaming 1 40minute tv show used more.. yes 40 minutes.

if people cannot handle expanding bitcoins network. then they need to complain to their internet providor and to tv show streaming services and tell them to stop offering such video material as its impossible to watch tv shows... oh wait it is possible..

i really find it foolish that in a era of online gaming and video streaming. that some core fangirls still live in the era of floppy disks.
even core devs have stopped supporting compatibility with vista/older systems. so subtly they are supporting the notion that systems can handle greater capacity by only supporting systems which by default are newer. so get your head out of the windows 98 era and update yourself, its 2019 not 1999

[pooya87]

those saying increasing the blocksize comes with a cost.. are outdated notions from years ago
its already been demythed about hard drive and internet costs.
bitcoin only uses under 200mb a day, yes a DAY (1.4mb * 144 blocks(most blocks are under 1.4mb))
just streaming 1 40minute tv show used more.. yes 40 minutes.

well if you compare bitcoin with a TV show then of course it seems trivial. but the problem is that when you stream a video you aren't validating the frames of the show! so your CPU doesn't do anything. it just receives and displays.
with bitcoin you have to receive and validate which means pressure on CPU and RAM to look up previous transactions in UTXO set and validate using EC calculations that are costly.

in short there is a cost to increasing the blocksize and the cost is not just hard disk or bandwidth related, it is also CPU and RAM related.
and the trick is to keep the cost manageable by majority. if you want to argue about this last part, that would be a different discussion but don't say there is no cost but storage and bandwidth.

[bob123]

those saying increasing the blocksize comes with a cost.. are outdated notions from years ago

Increasing the blocksize has nothing to do with scaling.

I mean.. of course you could increase the blocksize and accept all the downsides with it. But you won't solve any problems with it.
You will be confronted with the same issue again in X years. And what are you going to do then ? Increase it again ? To have even more negative effects ?

Relying on a bigger blocksize in order to scale for more transactions is a bad approach. What we need is real scaling, not postponing the problem while accepting negative effects.
There are much better approaches than the blocksize increase.

[DooMAD]

bitcoin only uses under 200mb a day, yes a DAY (1.4mb * 144 blocks(most blocks are under 1.4mb))
just streaming 1 40minute tv show used more.. yes 40 minutes.

Now run a streaming service where the users keep a copy of every show every user has ever watched in order to watch the latest episodes and tell me how few resources that uses.  

The cost is not the amount per day and nothing else.  The cost is the total current size of the blockchain plus the amount it increases by each day.  Unless you reconfigure the settings on your node, you need to broadcast information to bootstrap new nodes and keep other nodes up to date.  The recommended specification for upload allowance if you don't change the standard settings is:
 

5 GB/day (150 GB/month)

[franky1]

Now run a streaming service where the users keep a copy of every show every user has ever watched in order to watch the latest episodes and tell me how few resources that uses.  

users individual transactions are hundred kilobytes.. not 500mb+.. so your comparison is flawed

but as to a system that keeps a copy.. well that the temporary internet folder. think about it!

edit:

Now run a streaming service where the users keep a copy of every show every user has ever watched in order to watch the latest episodes and tell me how few resources that uses.  

users individual transactions are hundred kilobytes.. not 500mb+.. so your comparison is flawed

It's YOUR COMPARISON you fucking moron.  I know it's flawed.  That was the entire point.  

just streaming 1 40minute tv show used more.. HERPA DERP DERP

So yeah, it's not a valid comparison.  That's literally the first thing you've ever said that makes the slightest bit of sense.

doomad your comparing individual transactions to large blocks of data called tv episodes.

take for instance a 500mb 40minute show.... in comparison of time thats ~4 blocks... which is only ~5mb  yep bitcoin is 1% comparison to streaming.
as for saying 5gb a day... (facepalm) thats for users with a HIGH node count connection. if you dont want to use so much, simply come down to just a few nodes.

as for bob123's comment about someone postponing scaling... sorry but 2015 there was a massive scaling debate and a conference to try coming to an agreement. 4 years later.. yep 4 years later and scaling has not ben solved because core devs postponed scaling the btc network in favour of throwing people off the network to use other networks instead...(which is not scaling)

as for pooya87's comment about CPU... (mega facepalm) sorry dude but you might want to check the stats.
a 10yo CPU can handle over 1million hashes PER SECOND. so a block of ~2500 tx is childs play.
as for the ECDSA part..a block of 2500tx could validate quickly.. this faceppalm issue you raise was sorted 6+ years ago

lets even use LNs positive promotional material
after all channel signing in LN uses the same thing and guess what. LN advertises millions of tp/s... think about it.

[DooMAD]

Now run a streaming service where the users keep a copy of every show every user has ever watched in order to watch the latest episodes and tell me how few resources that uses.  

users individual transactions are hundred kilobytes.. not 500mb+.. so your comparison is flawed

It's YOUR COMPARISON you fucking moron.  I know it's flawed.  That was the entire point.  

just streaming 1 40minute tv show used more.. HERPA DERP DERP

So yeah, it's not a valid comparison.  That's literally the first thing you've ever said that makes the slightest bit of sense.

[franky1]

the risks of onchain scaling are not about hard drive sizes or CPU requirements. but instead the FOOLISH allowance of transactions to have upto 4000 signatures per tx. by simply reducing it to just a few hundred. still allows people to add multiple inputs. but removing the potential risk that was highlighted 5+ years ago about legacy verification/propogation delays

[pooya87]

as for pooya87's comment about CPU... (mega facepalm) sorry dude but you might want to check the stats.
a 10yo CPU can handle over 1million hashes PER SECOND. so a block of ~2500 tx is childs play.
as for the ECDSA part..a block of 2500tx could validate quickly.. this faceppalm issue you raise was sorted 6+ years ago

the point was that you were comparing "streaming a video" with "ECDSA verification". and you can't make that comparison.

lets even use LNs positive promotional material
after all channel signing in LN uses the same thing and guess what. LN advertises millions of tp/s... think about it.

on lightning network, unlike on main-net, not all the nodes are supposed to verify all the transactions! so even if a million tx/s happens, a single node may not handle (and verify) more than a hundred.

[franky1]

as for pooya87's comment about CPU... (mega facepalm) sorry dude but you might want to check the stats.
a 10yo CPU can handle over 1million hashes PER SECOND. so a block of ~2500 tx is childs play.
as for the ECDSA part..a block of 2500tx could validate quickly.. this faceppalm issue you raise was sorted 6+ years ago

the point was that you were comparing "streaming a video" with "ECDSA verification". and you can't make that comparison.

no the actual comment should you actually read and quote fully was about HARD DRIVE and INTERNET COSTS. no mention about cpu processing
here ll say it again
"its already been demythed about hard drive and internet costs.
bitcoin only uses under 200mb a day, yes a DAY (1.4mb * 144 blocks(most blocks are under 1.4mb))
just streaming 1 40minute tv show used more.. yes 40 minutes."

lets even use LNs positive promotional material
after all channel signing in LN uses the same thing and guess what. LN advertises millions of tp/s... think about it.

on lightning network, unlike on main-net, not all the nodes are supposed to verify all the transactions! so even if a million tx/s happens, a single node may not handle (and verify) more than a hundred.

dang seems your a flip flopper too.. do you realised if LN devs say a computer can signature verify X transactions a second.... then here is the point, wait for it, its a shocker... a computer can signature verify X transactions a second

saying that bitcoin is slow, redundant and faulty.. but then praising LN in a cloud of pink glory is the opposite of reality
if a node can signature verify X times s second in ln then it can do that same speed in bitcoin.

thats the point. pretending bitcoin takes too long to signature verify is propaganda and done by fools who actually hate the btc network, but those same fools love the devs that work hard to find ways to push people off the network by promoting other networks of less security and more flaws

[Wind_FURY]

franky1, you assume too much. You assume that everyone will accept more costs, and upgrade their hardware, their bandwidth, or both.

But if you accept a less-decentralized/scaled-in network that comes with bigger block sizes, then OK.

[franky1]

franky1, you assume too much. You assume that everyone will accept more costs, and upgrade their hardware, their bandwidth, or both.

But if you accept a less-decentralized/scaled-in network that comes with bigger block sizes, then OK.

are you that ignorant?
you assume that it cost alot to run a full node.
seriously, current btc can run on a raspberry pi which by default is MULTIPLE times worse than an AVERAGE desktop/laptop

this means that upgrading the network means people do not need to buy expensive equipment
people on average upgrade their computer every 2-5 years anyway... even without thinking about bitcoin.

thus just using recent generation computers and continuing normal upgrade trends, people will not see any extra expense that they would not have paid otherwise.

but i do feel sorry for you if you are still using a computer from 2009(or older). and refusing to upgrade, but if your having issues, with a slow computer.. maybe it is time you upgrade anyway.
if your computer freezes playing solitaire or opening a web browser. its time to upgrade


since 2017 btc has expanded by ~70gb a year. thats only 140gb-350gb (2-5 year respectively)
people can get 4tb hard drives these days
this is ~50 years current rules
this is ~25 years legacy 2mb
this is ~12 years legacy 4mb
this is ~6 years legacy 8mb

but here is another point. the home userbase do not need to know personally that every transaction verifies. all they care about is the ones involving them. so while home userbase will end up just using custodial services and side/alt networks(ln) which would be compatible with phone apps so they dont need to lug around a desktop/laptop to 'buy coffee'. it the merchants that do actually need to verify multiple transactions and have more desire to secure their funds, who will want to be full nodes. so its the merchant userbase that will be the main full node users.

dont believe me.. check it out. coinbase has millions of user accounts (millions of home users). yet 10k full nodes. this shows home users dont want or care to be full nodes

if a business doesnt have a computer.. then the business wont be successful in doing many things, like modern accounting, stock control. staff wages, customer relations/complaints.
so please dont try saying btc shouldnt evolve 'because of home users' when your not even taking into account real live situations, facts and stats

but if you want to remain in the myth that all home users should be full nodes then you have to accept if all million+ home users were full nodes right now. to propogate the latest block to all users (without changing blocksize) would take longer.

dont believe me. well find 3 friends and play 'pass the parcel' count how long it takes for the parcel to get to you. then get 3000 friends. play again. and count.
without needing to know the length of time to verify the person is holding the parcel. just passing it along takes longer.
think about it

also if you still want to argue that home users cant cope with btc. here is some tips:
1. your saying btc is broken and not viable (anti-btc to promote another network(ln))
2. go tell youtube, twitch, torrent users that the internet is broke and no one can up/download
3. go tell online gamers that they cant install large games and cant then play them because the internet is broke

or admit that the internet is fast, computers are capable and it does not require 'super computers'

that supercomputer requirement myth has been debunked ages ago. even by some of the core devs. so just drop that myth

[DooMAD]

*stuff*

Translation:

If 99% of users relied on SPV and only the big corporations ran full nodes, franky1 could realise his dream of near-free transactions because other people would then be paying the underlying cost for it.  But sadly can't reconcile this with the reality of real users wanting to avoid trusting a few big corporations to use Bitcoin.  So the users keep running nodes that enforce small-ish blocks.  Even when the users agree to allow slightly more throughput, it's never good enough for franky1.  Hence never getting what he wants and flailing like an entitled infant about how all the other Bitcoin users don't understand his supposed genius.

It's not rocket science.  It's entirely possible for you to have the network you want.  But it's clearly not this network.  The users on this network do not want your glorious vision of cheap and centralised.  The things you desire are not even remotely compatible with what we're doing.  How is this not obvious to you?  There are other networks that cater to your desires, but you're still hellbent on turning this network into those other networks, even though that's completely redundant and the users on this chain disagree with the approach those other networks are taking.  How do you think you can be a constructive and helpful part of this network when you fundamentally disagree with the software 95+% of the users are running?  And then you have the gall, the unashamed audacity and barefaced cheek, to claim that we don't understand consensus?  What's wrong with your brain?  Seriously?   

[Wind_FURY]

franky1, you assume too much. You assume that everyone will accept more costs, and upgrade their hardware, their bandwidth, or both.

But if you accept a less-decentralized/scaled-in network that comes with bigger block sizes, then OK.

are you that ignorant?

Relax.

you assume that it cost alot to run a full node.
seriously, current btc can run on a raspberry pi which by default is MULTIPLE times worse than an AVERAGE desktop/laptop

Because the block size is regulated to be small. Plus bandwidth is more of an issue in running a full node than hardware requirements.

dont believe me.. check it out. coinbase has millions of user accounts (millions of home users). yet 10k full nodes. this shows home users dont want or care to be full nodes

That doesn't mean it's ok to take away the users' ability to run one by making costs higher.

but if you want to remain in the myth that all home users should be full nodes then you have to accept if all million+ home users were full nodes right now. to propogate the latest block to all users (without changing blocksize) would take longer.

Why is it a myth? The miners have taken away the users' ability to mine through cartelization, now you want users to give up their ability to run full nodes too? The network will not scale out, it will centralize.

[aliashraf]

*stuff*

It's not rocket science.  It's entirely possible for you to have the network you want.  But it's clearly not this network.  The users on this network do not want your glorious vision of cheap and centralised.  The things you desire are not even remotely compatible with what we're doing.  How is this not obvious to you?  There are other networks that cater to your desires, but you're still hellbent on turning this network into those other networks, even though that's completely redundant and the users on this chain disagree with the approach those other networks are taking.  How do you think you can be a constructive and helpful part of this network when you fundamentally disagree with the software 95+% of the users are running?  And then you have the gall, the unashamed audacity and barefaced cheek, to claim that we don't understand consensus?  What's wrong with your brain?  Seriously?    

I'm absolutely against these arguments. Basically, you are saying bitcoin is what it is and anybody who doesn't like it better just leave! So, what's the point of reading, thinking, discussing or any other intellectual practice in the bitcoin community, heh? Let's just buy a few coins and wait for bulls to show up and we become rich?  

It is not how the technology works bro. Read something.

I do agree that governance is a crisis in bitcoin and things are complicated when it comes to serious problems such as scaling and centralization problems but it doesn't imply any kind of dead-end or logical paradox. There are always possibilities and options. Let's elaborate a bit more:

I hate to say this, but sidechains show us a way out of this nightmare. I hate saying this because the name, sidechain, has been totally alienated from its original notion by some people to refer to a stupid, 2-way pegged, off-chain solution which I think is a fruitless idea. Who in the hell would freeze his bitcoins in an experimental chain to reclaim it later? And more importantly, how would it help with an already centralized mining sphere? But if we could get rid of this "2-way pegged" shit, a sidechain would be a great idea.

Suppose we have a side chain with great scaling and decentralization (both, for Buterine's information) properties which competes with legacy chain fairly and nondisruptively while it is using the same hash power of bitcoin network through merged mining. How is that?

And yet it is one of the many ways out there waiting for people to explore them. So, it is better to focus on the technical stuff instead of suppressing discussions and pushing people out of the bitcoin community, in the name of conservatism.

[DooMAD]

*stuff*

It's not rocket science.  It's entirely possible for you to have the network you want.  But it's clearly not this network.  The users on this network do not want your glorious vision of cheap and centralised.  The things you desire are not even remotely compatible with what we're doing.  How is this not obvious to you?  There are other networks that cater to your desires, but you're still hellbent on turning this network into those other networks, even though that's completely redundant and the users on this chain disagree with the approach those other networks are taking.  How do you think you can be a constructive and helpful part of this network when you fundamentally disagree with the software 95+% of the users are running?  And then you have the gall, the unashamed audacity and barefaced cheek, to claim that we don't understand consensus?  What's wrong with your brain?  Seriously?    

I'm absolutely against these arguments. Basically, you are saying bitcoin is what it is and anybody who doesn't like it better just leave! So, what's the point of reading, thinking, discussing or any other intellectual practice in the bitcoin community, heh? Let's just buy a few coins and wait for bulls to show up and we become rich?  

It is not how the technology works bro. Read something.

Off-topic:  Good timing, how do you feel about Carlton's assertion that you're "very likely anonymint"?

On topic:  The technology also doesn't work in a way where <5% of the network can dictate terms to the other 95%.  Regardless of your personal feelings towards how I've phrased it, people can run whatever software they like, but it doesn't mean they can force their proposed rule changes upon an unwilling network. If enough people on the network elect to enforce new rules, fair enough.  But at some point, when it becomes abundantly clear that you are very much in the minority, there comes a time where it would behoove you to recognise that pissing into the wind isn't serving your goals very well.

And, for what it's worth, I'm intrigued by the potential for sidechains.  If they can provide some form of compromise that keeps all sides reasonably satisfied, it will have my support.

[aliashraf]

*stuff*

It's not rocket science.  It's entirely possible for you to have the network you want.  But it's clearly not this network.  The users on this network do not want your glorious vision of cheap and centralised.  The things you desire are not even remotely compatible with what we're doing.  How is this not obvious to you?  There are other networks that cater to your desires, but you're still hellbent on turning this network into those other networks, even though that's completely redundant and the users on this chain disagree with the approach those other networks are taking.  How do you think you can be a constructive and helpful part of this network when you fundamentally disagree with the software 95+% of the users are running?  And then you have the gall, the unashamed audacity and barefaced cheek, to claim that we don't understand consensus?  What's wrong with your brain?  Seriously?    

I'm absolutely against these arguments. Basically, you are saying bitcoin is what it is and anybody who doesn't like it better just leave! So, what's the point of reading, thinking, discussing or any other intellectual practice in the bitcoin community, heh? Let's just buy a few coins and wait for bulls to show up and we become rich?  

It is not how the technology works bro. Read something.

Off-topic:  Good timing, how do you feel about Carlton's assertion that you're "very likely anonymint"?

I liked that guy his ban was unfair but I accept the fact that he went too far in his idea regarding anti-segwit coup d'état.
Although we used to share some kind of dissatisfaction with current bitcoin, we are not the same person:
anonymint was a warrior, I'm a developer;
he was fighting against core devs I'm trying to convince them by my upcoming codes;
anonymint was a US southern citizen, I was born in Iran, never left Iran and have no plan to do so.
....

That all said, It is an honor for me to be somehow, somewhat mistaken with that guy, he was an exceptionally original thinker, unlike Carlton.  

On topic:  The technology also doesn't work in a way where <5% of the network can dictate terms to the other 95%.  Regardless of your personal feelings towards how I've phrased it, people can run whatever software they like, but it doesn't mean they can force their proposed rule changes upon an unwilling network. If enough people on the network elect to enforce new rules, fair enough.  But at some point, when it becomes abundantly clear that you are very much in the minority, there comes a time where it would behoove you to recognise that pissing into the wind isn't serving your goals very well.

Fair enough. But minority should have a voice anyway. We need to listen to each other and leave behind the bitter days of bcash fork and faketoshi. The crusade is over it is time to think radically and freely about anything that bitcoin needs to overcome its limitations.

And, for what it's worth, I'm intrigued by the potential for sidechains.  If they can provide some form of compromise that keeps all sides reasonably satisfied, it will have my support.

Good, good.

[DooMAD]

We need to listen to each other and leave behind the bitter days of bcash fork and faketoshi.

We'd love to do that, believe me, but some of the very users in this topic keep regurgitating faketoshi's nauseating false arguments and lies.  If we could consign that nonsense to the history books and move forward, that would be great.  But it seems some people can't help but drag their heels and live in the past, namely the period of time prior to SegWit being activated.  I'll start listening if they ever get with the times.