ETH doesn't scale. This is not surprising or new. Then they use secondary chains to batch transactions so they can backup on chain less often.
What seems to be new here is that these secondary chains that already exist are getting integrated into ETH while still being a separate chain.
So after reinventing a database they are implementing sharding on top of it which to me seems like trailing already walked paths that will guide you nowhere new.
The result is a less flexible and more complex system that attenuates but doesn't solve the scaling issue.
> This is how any other peer-to-peer network (eg. BitTorrent) works too, so surely we could make blockchains work the same way
> For many years, it has been common for some high-performance chains to claim that they solve the trilemma without doing anything clever at a fundamental architecture level, typically by using software engineering tricks to optimize the node. This is always misleading, and running a node in such chains always ends up far more difficult than in Ethereum.
Not saying you're wrong, but I think if you disagree with a 5k+ word article full of careful technical explanations and diagrams (I am not an expert on Ethereum and really appreciated how the author explained things in a way that taught me stuff) then you need to put in a bit more effort than just naming the challenges again.
Crypto maxis like above really are the worst part of the entire crypto space. They go around spewing nonsense about whatever chain they perceive as "best". It's rather toxic and probably pretty bad for their mental health.
It's always kinda fascinating too. Like the person saw a post about Ethereum on here and raged out and created an account just to spew their digital gospel.
They don't pregonate what they perceive is best, or try to evangelize.
They hold the asset that they recommend, so that it increases in price.
This dispels another common misconception that shills get paid, nope, they just buy into an asset at some early point in the early to late spectrum of a ponzi scheme, and then try to sell it by recommendation.
No, the batching is done so that the transactions form a data structure that's amenable to optimistic verification or succinct zero knowledge proofs. All transactions can be fully verified onchain, every block, and Ethereum rollup transaction flows would see generate massive scalability boost ups.
The incredible thing about the rollup scaling roadmap is that transactions across a rollup can be composable with each other while using different shards.
This means Ethereum can massively scale with no compromises on decentralization or composability.
Instead of paying 10 bucks to pollute the main Ethereum blockchain, I pay pennies in some second hand pepsi wannabe blockchain like Tron or Solana. I trade my assets once to the secondary chain and then I microtransact on what's effectively an L2 network.
Same thing with banks, we can't have dozens of transactions per day for stuff like a hamburger and a coffee, so they invented credit cards as a sort of L2 layer that does smaller transactions.
And then wallets like paypal popped up to fill the void in places where the transaction was too low or insecure to even warrant the use of a card, like when you buy porn o dumb videogames. So L3.
Having everything on THE blockchain is not actually the decentralized vision we would have in mind, quite the opposite, there's a bunch of blockchains and they communicate with each other.
> Same thing with banks, we can't have dozens of transactions per day for stuff like a hamburger and a coffee
We can't? I guess this is US specific, because I've been able to use my bank card for all transactions since I have a bank account (at least 20 years). All transactions hit the bank account directly and are rejected if balance is insufficient.
I don't think credit cards have anything to do with scaling.
Depends on what level of the banking system you’re talking. Only other banks can transact with the national bank, so you can’t pay for your hamburgers there.
When I pay for public transportation, they batch my travels per day to minimise the transaction fees because they figured most people travel twice per day.
The public transport transaction fees are because VISA/Mastercard. Atleast in europe bank accounts transact wihtout fees. Also its becoming instant and in some countries even mandatory to be instant. Because of that QRcodes for transactions are picking up with many ecommerce stores pushing QR payment (simply bank transfer) because of the fees. Soon we might soon see VISA/Mastercard less banking.
That's actually a bit disappointing. Someone I know recently had a QR code sent to them in the mail, which they scanned out of curiosity -- not a technical person. Turns out it was a URL to a rather malicious server and they had to scramble to secure logins and such. Encouraging people to scan QR codes as if stuff like that can't happen just makes it more likely for stuff like that to happen.
Not to say that it's strictly a bad thing, just that this particular implementation detail could be improved.
> What were they running, Windows XP with 2-year old chrome browser?
Like many consumers: iOS with automatic updates turned on. It’s not news, either; there are groups who discover and use 0-day exploits in order to allow their customers -- usually a government -- to infiltrate a target smart phone -- by phone number, no less -- and install spyware on it.
NSO Group in particular made the news in the past few years because their software was used against the devices of US government employees, which goes against the contract they had/have with the US government. Famously, a journalist was targeted with their software and subsequently gunned down at a gas station, naturally because they had their phone on them, which was used as a real-time location tracker by the assassination group.
Luckily, most people aren’t investigative journalists so there’s a small worry about assassination. Unluckily, there’s a lot of global organized crime, which ends up paying well for those who do impactful work... like perhaps finding a Safari/iOS vulnerability or a camera/iOS vulnerability and exploiting it to exfiltrate bank account credentials.
If you get hit by Apple 0-day and infected with spyware, what good does changing the passwords on your accounts do? The new passwords would be immediately stolen anyway, unless you throw away the phone and get a new one - which seems pretty excessive reaction to opening a single web page, no matter how malicious.
Worth noting that credit cards precede debit cards. In effect credit cards were Layer 2 until the Layer 1 adapted and managed to scale and implement the features from L2. Like Bitcoin with Segwit and lightning network, or what Vitalik is proposing here.
Credit cards preceded the Internet. The problem wasn't that banks didn't scale, it was that the latency of informing the bank of a transaction was prohibitive.
Unfortunately, blockchains sprung into popularity right around the time traditional bank transfers were finally prodded into the 21st century (SEPA, UPI, FedNow, etc).
Which means they're now competing against a digital, trusted-counterparty system.
That is both a cynical take and an anachronistic one.
Sure you can argue that currently it is only used for get rich quick schemes.
But to argue that it was get rich quick schemes from the start is innecessary.
Also while get rich quick schemes are a problem with crypto, and there are crypto securities that evade regulation, there's also tax evasion, cybersecurity ransomware payments, drug trade, and not an insignificant amount of real legitimate transactions.
You can again argue that the real transactions serve to pool illegitimate usecases with legitimate ones, like an xkcd mirror on the Tor network.
But that would be a more nuanced and accurate take.
Blockchains compete with the existing financial system, which is largely held together by the major central banks and commercial banks.
And because of poor timing as mentioned, they just seem like a shabby deal since nowadays it’s entirely possible to get nearly all the benefits in the existing system without the drawbacks.
That is so true but funny in light of the fact that Bitcoin does (did) scale perfectly well to Visa-like levels of transactions. The ONLY problem with Bitcoin scaling is the for profit company, Blockstream, having hijacked the GitHub repo and injected hacks like RBF and Segwit.
The original Bitcoin (Bitcoin Cash) still works and scales totally fine.
Your words are interesting, but they need a strong source of truth to be relevant. Would you please like to add an URL that supports your thesis? Thanks!
Bitcoin currently is just incredibly unfriendly. On any exchange wallet, you now have multiple Bitcoin addresses - segwit, lightning, bitcoin, bep20/erc20, a separate taproot address, separate ordinals address
its honestly intimidating
this thing needs to be simplified, not made more complex
the stupid ordinals/runes stuff has made it even more complex
I like it when L1 blockchains just increase their system requirements over time, acknowledging improvements in global bandwidth like everything else
I like how we don’t have to debate blocksize on other chains
Someone looked at the Bitcoin whitepaper and said, hey what if nodes required 256gb RAM and a Threadripper CPU because it’s not 2008 anymore. And it was awesome
I would like to see an EVM take a similar approach
On the other hand, I think technology should get more efficient and better over time, rather than expand to be maximally inefficient. This is probably why I dislike blockchain.
The issue is not so much hardware requirements (although they matter as well) but bandwidth which is the main limitation to a distributed decentralized network that's not hosted in datacenters
> Someone looked at the Bitcoin whitepaper and said, hey what if nodes required 256gb RAM and a Threadripper CPU because it’s not 2008 anymore.
It's a very contentious issue in the Bitcoin world but Satoshi himself wrote code, that was commented out, showing that he envisioned bigger blocks at some point.
Doubling the size of the blocks wouldn't have been an issue.
And politics got in the way and shady deals were made: IIRC everybody agreed on adding "segwit" and 2x bigger blocks to Bitcoin. But as soon as Segwit was locked in, a majority said "oh after all screw you 2x bigger block, we're not doing that". They knew all along they'd act like snakes but first made sure enough people voted to lock segwit in.
That's when Roger Ver (aka "Bitcoin Jesus", who owns the bitcoin.com domain) said "screw you guys, I'm doubling the size of the blocks like we all agreed to and like Satoshi envisioned" and the Bitcoin vs Bitcoin Cash fork happened.
It's not as if Satoshi didn't plan for bigger blocks at some point: he had both code and comments on forums explaining that at some point the block size should grow.
But politics got in the way and a little cabal of "faketoshis" made sure bigger blocks couldn't happen.
P.S: Bitcoin Cash is a failure but, arguably, it is Satoshi's original vision, where blocks got bigger as machines got bigger.
Cardano uses PoS and is therefore inferior. All Proof of Stake chains have been a downward spiral because existing holders have no incentive to contribute to the economy (hashpower/decentralization, commerce or innovation). ETH is starting to follow the same path and stagnate.
eUTXO is Turing complete. Ergo, Cardano and Ethereum are all Turing complete unlike plain UTXO. This means they will have a much harder time figuring out the state. UTXO chains only have to look up the TX inputs to maintain state so they can scale much faster.
ETH doesn't scale. This is not surprising or new. Then they use secondary chains to batch transactions so they can backup on chain less often.
What seems to be new here is that these secondary chains that already exist are getting integrated into ETH while still being a separate chain.
So after reinventing a database they are implementing sharding on top of it which to me seems like trailing already walked paths that will guide you nowhere new.
The result is a less flexible and more complex system that attenuates but doesn't solve the scaling issue.
Yes TFA says largely the same
> This is how any other peer-to-peer network (eg. BitTorrent) works too, so surely we could make blockchains work the same way
> For many years, it has been common for some high-performance chains to claim that they solve the trilemma without doing anything clever at a fundamental architecture level, typically by using software engineering tricks to optimize the node. This is always misleading, and running a node in such chains always ends up far more difficult than in Ethereum.
Not saying you're wrong, but I think if you disagree with a 5k+ word article full of careful technical explanations and diagrams (I am not an expert on Ethereum and really appreciated how the author explained things in a way that taught me stuff) then you need to put in a bit more effort than just naming the challenges again.
[flagged]
ok shill
Crypto maxis like above really are the worst part of the entire crypto space. They go around spewing nonsense about whatever chain they perceive as "best". It's rather toxic and probably pretty bad for their mental health.
It's always kinda fascinating too. Like the person saw a post about Ethereum on here and raged out and created an account just to spew their digital gospel.
They don't pregonate what they perceive is best, or try to evangelize.
They hold the asset that they recommend, so that it increases in price.
This dispels another common misconception that shills get paid, nope, they just buy into an asset at some early point in the early to late spectrum of a ponzi scheme, and then try to sell it by recommendation.
Just another ponzi
No, the batching is done so that the transactions form a data structure that's amenable to optimistic verification or succinct zero knowledge proofs. All transactions can be fully verified onchain, every block, and Ethereum rollup transaction flows would see generate massive scalability boost ups.
The incredible thing about the rollup scaling roadmap is that transactions across a rollup can be composable with each other while using different shards.
This means Ethereum can massively scale with no compromises on decentralization or composability.
Which is quite decentralized in a way.
Instead of paying 10 bucks to pollute the main Ethereum blockchain, I pay pennies in some second hand pepsi wannabe blockchain like Tron or Solana. I trade my assets once to the secondary chain and then I microtransact on what's effectively an L2 network.
Same thing with banks, we can't have dozens of transactions per day for stuff like a hamburger and a coffee, so they invented credit cards as a sort of L2 layer that does smaller transactions.
And then wallets like paypal popped up to fill the void in places where the transaction was too low or insecure to even warrant the use of a card, like when you buy porn o dumb videogames. So L3.
Having everything on THE blockchain is not actually the decentralized vision we would have in mind, quite the opposite, there's a bunch of blockchains and they communicate with each other.
> Same thing with banks, we can't have dozens of transactions per day for stuff like a hamburger and a coffee
We can't? I guess this is US specific, because I've been able to use my bank card for all transactions since I have a bank account (at least 20 years). All transactions hit the bank account directly and are rejected if balance is insufficient.
I don't think credit cards have anything to do with scaling.
Depends on what level of the banking system you’re talking. Only other banks can transact with the national bank, so you can’t pay for your hamburgers there.
When I pay for public transportation, they batch my travels per day to minimise the transaction fees because they figured most people travel twice per day.
The public transport transaction fees are because VISA/Mastercard. Atleast in europe bank accounts transact wihtout fees. Also its becoming instant and in some countries even mandatory to be instant. Because of that QRcodes for transactions are picking up with many ecommerce stores pushing QR payment (simply bank transfer) because of the fees. Soon we might soon see VISA/Mastercard less banking.
> QRcodes
That's actually a bit disappointing. Someone I know recently had a QR code sent to them in the mail, which they scanned out of curiosity -- not a technical person. Turns out it was a URL to a rather malicious server and they had to scramble to secure logins and such. Encouraging people to scan QR codes as if stuff like that can't happen just makes it more likely for stuff like that to happen.
Not to say that it's strictly a bad thing, just that this particular implementation detail could be improved.
this is insane... one malicious page visit and person "had to scramble to secure logins" (I assume this means "change passwords?")
What were they running, Windows XP with 2-year old chrome browser?
> What were they running, Windows XP with 2-year old chrome browser?
Like many consumers: iOS with automatic updates turned on. It’s not news, either; there are groups who discover and use 0-day exploits in order to allow their customers -- usually a government -- to infiltrate a target smart phone -- by phone number, no less -- and install spyware on it.
NSO Group in particular made the news in the past few years because their software was used against the devices of US government employees, which goes against the contract they had/have with the US government. Famously, a journalist was targeted with their software and subsequently gunned down at a gas station, naturally because they had their phone on them, which was used as a real-time location tracker by the assassination group.
Luckily, most people aren’t investigative journalists so there’s a small worry about assassination. Unluckily, there’s a lot of global organized crime, which ends up paying well for those who do impactful work... like perhaps finding a Safari/iOS vulnerability or a camera/iOS vulnerability and exploiting it to exfiltrate bank account credentials.
If you get hit by Apple 0-day and infected with spyware, what good does changing the passwords on your accounts do? The new passwords would be immediately stolen anyway, unless you throw away the phone and get a new one - which seems pretty excessive reaction to opening a single web page, no matter how malicious.
Worth noting that credit cards precede debit cards. In effect credit cards were Layer 2 until the Layer 1 adapted and managed to scale and implement the features from L2. Like Bitcoin with Segwit and lightning network, or what Vitalik is proposing here.
Credit cards preceded the Internet. The problem wasn't that banks didn't scale, it was that the latency of informing the bank of a transaction was prohibitive.
Unfortunately, blockchains sprung into popularity right around the time traditional bank transfers were finally prodded into the 21st century (SEPA, UPI, FedNow, etc).
Which means they're now competing against a digital, trusted-counterparty system.
Blockchains don't compete with banks, they compete with stock markets and other exchanges.
Basically Bitcoin exists because nerds wanted to play Wolf of Wall Street and thought the government was holding them down.
Bitcoin was literally created as a P2P replacement for the financial industry.
The title of its white paper is “Bitcoin: A Peer-to-Peer Electronic Cash System”[0].
How is that not ‘competing with banks’?
[0] https://bitcoin.org/bitcoin.pdf
> Bitcoin was literally created as a P2P replacement for the financial industry.
No, that was "literally" the user-facing justification.
But it was created as a get-rich-quick scheme for nerds.
That is both a cynical take and an anachronistic one.
Sure you can argue that currently it is only used for get rich quick schemes.
But to argue that it was get rich quick schemes from the start is innecessary.
Also while get rich quick schemes are a problem with crypto, and there are crypto securities that evade regulation, there's also tax evasion, cybersecurity ransomware payments, drug trade, and not an insignificant amount of real legitimate transactions.
You can again argue that the real transactions serve to pool illegitimate usecases with legitimate ones, like an xkcd mirror on the Tor network.
But that would be a more nuanced and accurate take.
Blockchains compete with the existing financial system, which is largely held together by the major central banks and commercial banks.
And because of poor timing as mentioned, they just seem like a shabby deal since nowadays it’s entirely possible to get nearly all the benefits in the existing system without the drawbacks.
> Blockchains compete with the existing financial system
Nobody actually uses blockchains for accounts or money transfers. (Nobody as in actually nobody.)
Lots of people use blockchains for speculation purposes, though.
Who said the existing financial system is limited to "accounts or money transfers"?
Wrong. Usd Stablecoins (tether) are very popular in my country. Salaries are paid with that shit
The people who say nobody uses blockchains for actual money are showing their privilege.
Nano is one crypto that is a block-lattice instead of a blockchain and doesn't need the nested layering you refer to.
>ETH doesn't scale
That is so true but funny in light of the fact that Bitcoin does (did) scale perfectly well to Visa-like levels of transactions. The ONLY problem with Bitcoin scaling is the for profit company, Blockstream, having hijacked the GitHub repo and injected hacks like RBF and Segwit.
The original Bitcoin (Bitcoin Cash) still works and scales totally fine.
> Visa-like
is that the 1950s visa like or the 1960s visa like transaction levels?
Your words are interesting, but they need a strong source of truth to be relevant. Would you please like to add an URL that supports your thesis? Thanks!
> Bitcoin Cash
Please… let’s not all get Flat Earth around here
No, really... Flat Earth would be BSV.
BCH is more like JW "truthers" or something.
Sedevacantists
Pretty sure bitcoin's most well known sedevacantist, luke-jr, sided with bitcoin not bitcoin cash lol.
Bitcoin currently is just incredibly unfriendly. On any exchange wallet, you now have multiple Bitcoin addresses - segwit, lightning, bitcoin, bep20/erc20, a separate taproot address, separate ordinals address
its honestly intimidating
this thing needs to be simplified, not made more complex
the stupid ordinals/runes stuff has made it even more complex
Why do you think that it doesn't solve the scaling issue?
I like it when L1 blockchains just increase their system requirements over time, acknowledging improvements in global bandwidth like everything else
I like how we don’t have to debate blocksize on other chains
Someone looked at the Bitcoin whitepaper and said, hey what if nodes required 256gb RAM and a Threadripper CPU because it’s not 2008 anymore. And it was awesome
I would like to see an EVM take a similar approach
On the other hand, I think technology should get more efficient and better over time, rather than expand to be maximally inefficient. This is probably why I dislike blockchain.
At least ETH abandoned proof-of-waste.
But they also do that
The issue is not so much hardware requirements (although they matter as well) but bandwidth which is the main limitation to a distributed decentralized network that's not hosted in datacenters
Did you skip the first sentence
> acknowledging improvements in global bandwidth like everything else
p2p still capped at sub 5mbps for most people
> Someone looked at the Bitcoin whitepaper and said, hey what if nodes required 256gb RAM and a Threadripper CPU because it’s not 2008 anymore.
It's a very contentious issue in the Bitcoin world but Satoshi himself wrote code, that was commented out, showing that he envisioned bigger blocks at some point.
Doubling the size of the blocks wouldn't have been an issue.
And politics got in the way and shady deals were made: IIRC everybody agreed on adding "segwit" and 2x bigger blocks to Bitcoin. But as soon as Segwit was locked in, a majority said "oh after all screw you 2x bigger block, we're not doing that". They knew all along they'd act like snakes but first made sure enough people voted to lock segwit in.
That's when Roger Ver (aka "Bitcoin Jesus", who owns the bitcoin.com domain) said "screw you guys, I'm doubling the size of the blocks like we all agreed to and like Satoshi envisioned" and the Bitcoin vs Bitcoin Cash fork happened.
It's not as if Satoshi didn't plan for bigger blocks at some point: he had both code and comments on forums explaining that at some point the block size should grow.
But politics got in the way and a little cabal of "faketoshis" made sure bigger blocks couldn't happen.
P.S: Bitcoin Cash is a failure but, arguably, it is Satoshi's original vision, where blocks got bigger as machines got bigger.
[flagged]
[flagged]
I believe Cardano also uses UTXO.
Cardano uses PoS and is therefore inferior. All Proof of Stake chains have been a downward spiral because existing holders have no incentive to contribute to the economy (hashpower/decentralization, commerce or innovation). ETH is starting to follow the same path and stagnate.
What about Ergo? It’s both eUTXO and PoW.
eUTXO is Turing complete. Ergo, Cardano and Ethereum are all Turing complete unlike plain UTXO. This means they will have a much harder time figuring out the state. UTXO chains only have to look up the TX inputs to maintain state so they can scale much faster.
How many UTXO do you hold?
Stagnate in what sense?
Why? Please explain and add some sources that support your thesis, thanks!
In my opinion we should focus on Xc.v4 running over Z1 hyper branches.
Utxo is the original blockchain primitive.