"Users don't care about decentralization" is another psyop you'll see, and we need to address that myth head on. So, we'll have a good Sunday TED talk to round off another day of Ethereum educational defense.
Users don't come out to say "I care about decentralization", but you can expect they won't be happy if something bad happened because the chain wasn't sufficiently decentralized and secure (I can think of some nice safety/liveness failures, for example). For context, I've done enough mythbusting to show that you can't just point at validator counts to show decentralization, especially when you have more factors at play (e.g., potential stake centralization from implementing a validator subsidy).
Again, it's true that 99% of the time nothing bad happens to even the most centralized of blockchains: no one has executed a safety attack that steals money from BNB users even though you have just 21 validators (21!) validators running a hybrid of Delegated Proof of Stake (DPoS) and Proof of Authority (PoA)--at least not yet. And BNB has ~$6b in TVL currently.
But BNB validators also halted the chain for eight hours () likely because you-know-who said "losing credibility is better than us to be having to deal with the fallout of a hack". Now, I'm not saying an emergency procedure to mitigate a massive exploit is not useful. But in crypto, you also have to think about what could go wrong--if it was that easy to get people to halt the chain, how easy might it be to actually finalize an invalid state transition?
"Someone else can run the chain" doesn't work here because it's not like exchanges, applications, and fiat on ramps/off-ramps are known to side with the minority in a contentious fork. So, you absolutely need to be sure that your "honest majority" has incentives to act honestly--and that you make it hard as possible for a dishonest minority to end up wielding a disproportionate amount of influence on the network.
Why would you need this? Well, no one is waiting to hand users back their money if the chain becomes insecure and shady validators (acting alone or in concert with shadowy super coders) exploit applications living on the chain. The SEC won't order a rollback of the chain to give people back their money DAO hack-style (and the possibility that Ethereum does a repeat of the post-DAO hack rollback is 0.0000).
Security experts tell you that security is asymmetric warfare--you need to be right 100% of the time to win, but your opponent needs to be right just *one* time to win and invalidate all the hard work you've done previously. It's no different with blockchains: it takes just one Black Swan event to get rekt (cf. Luna), even if you need to defend against the possibility of getting rekt all the time to remain secure.
Overfixating on decentralization and security isn't dumb. But it looks dumb because users don't have "available information" that shows the importance of a decentralized/secure chain--you don't get to experience the benefits of decentralization on a day-to-day basis. Chain failures are a Black Swan event, a zero day event.
In contrast, you have more available information about the benefits of scalability--you can see how fast the transactions are and how snappy the UX is. You're paying sub-cent ($0.00001) for fees, and it feels good not to pay $50 for a transaction on L1 or $0.01 on L2 ("[Insert favorite L1] has spoiled me. 😭❤️")
If I come up to you preaching the gospel of decentralization, I might be no different from a scientist telling you to avoid exposure to COVID-19--even though your immediate needs take priority over whatever I'm asking you to do. For example, I'm saying to avoid parties when the pleasure from YOLOing and partying feels more concrete compared to the (currently vague) pain I say you'll feel if you come down with the rona.
Avoiding Coronavirus is definitely good, especially if you have a weak immune system--you're literally at risk at dying due to a positive diagnosis, and that's as bad as Black Swan events get. If you knew in advance that you died because of one (and couldn't attend more parties), would you skip the party or go ahead to attend?
The imaginary individual in this little story can easily be the modern-day crypto user. I could very well say this person doesn't care about COVID-19 and only wants to party--it'd be true to an extent.
But it misses the fact that people might be hedonistic, but the instinct for long-term self-preservation runs even deeper than the desire for instant gratification. If I showed the individual a video of them dying because they attended just *one* party, they'd be (rightfully) scared and tolerate the short-term discomfort of skipping parties in exchange for getting to attend more parties in future.
I can also show you that chains *can* be centralized and insecure as a result, and that meant you couldn't build an application with the confidence that "it's never exploited or halted as a result of a chain failure". Or I can show you that, as a user, you can be censored indefinitely (have your funds frozen), and that safety attacks from not having a robust consensus layer can devalue your assets or outrightly steal them.
Obviously, you're going to be thinking twice about YOLOing into the next L1 that's offering buzzy apps, snappy UX, and fees so low you need to think up a new word for "lower than ultra-low" (whilst conveniently hiding all the subtle layers of centralization) after you become aware of these details. That's where "users don't care about decentralization" theory starts to break down pretty fast (we saw this happen when people moved from sidechains to L2s early into the rollup-centric roadmap).rekt.news/bnb-bridge-rek…
Users don't come out to say "I care about decentralization", but you can expect they won't be happy if something bad happened because the chain wasn't sufficiently decentralized and secure (I can think of some nice safety/liveness failures, for example). For context, I've done enough mythbusting to show that you can't just point at validator counts to show decentralization, especially when you have more factors at play (e.g., potential stake centralization from implementing a validator subsidy).
Again, it's true that 99% of the time nothing bad happens to even the most centralized of blockchains: no one has executed a safety attack that steals money from BNB users even though you have just 21 validators (21!) validators running a hybrid of Delegated Proof of Stake (DPoS) and Proof of Authority (PoA)--at least not yet. And BNB has ~$6b in TVL currently.
But BNB validators also halted the chain for eight hours () likely because you-know-who said "losing credibility is better than us to be having to deal with the fallout of a hack". Now, I'm not saying an emergency procedure to mitigate a massive exploit is not useful. But in crypto, you also have to think about what could go wrong--if it was that easy to get people to halt the chain, how easy might it be to actually finalize an invalid state transition?
"Someone else can run the chain" doesn't work here because it's not like exchanges, applications, and fiat on ramps/off-ramps are known to side with the minority in a contentious fork. So, you absolutely need to be sure that your "honest majority" has incentives to act honestly--and that you make it hard as possible for a dishonest minority to end up wielding a disproportionate amount of influence on the network.
Why would you need this? Well, no one is waiting to hand users back their money if the chain becomes insecure and shady validators (acting alone or in concert with shadowy super coders) exploit applications living on the chain. The SEC won't order a rollback of the chain to give people back their money DAO hack-style (and the possibility that Ethereum does a repeat of the post-DAO hack rollback is 0.0000).
Security experts tell you that security is asymmetric warfare--you need to be right 100% of the time to win, but your opponent needs to be right just *one* time to win and invalidate all the hard work you've done previously. It's no different with blockchains: it takes just one Black Swan event to get rekt (cf. Luna), even if you need to defend against the possibility of getting rekt all the time to remain secure.
Overfixating on decentralization and security isn't dumb. But it looks dumb because users don't have "available information" that shows the importance of a decentralized/secure chain--you don't get to experience the benefits of decentralization on a day-to-day basis. Chain failures are a Black Swan event, a zero day event.
In contrast, you have more available information about the benefits of scalability--you can see how fast the transactions are and how snappy the UX is. You're paying sub-cent ($0.00001) for fees, and it feels good not to pay $50 for a transaction on L1 or $0.01 on L2 ("[Insert favorite L1] has spoiled me. 😭❤️")
If I come up to you preaching the gospel of decentralization, I might be no different from a scientist telling you to avoid exposure to COVID-19--even though your immediate needs take priority over whatever I'm asking you to do. For example, I'm saying to avoid parties when the pleasure from YOLOing and partying feels more concrete compared to the (currently vague) pain I say you'll feel if you come down with the rona.
Avoiding Coronavirus is definitely good, especially if you have a weak immune system--you're literally at risk at dying due to a positive diagnosis, and that's as bad as Black Swan events get. If you knew in advance that you died because of one (and couldn't attend more parties), would you skip the party or go ahead to attend?
The imaginary individual in this little story can easily be the modern-day crypto user. I could very well say this person doesn't care about COVID-19 and only wants to party--it'd be true to an extent.
But it misses the fact that people might be hedonistic, but the instinct for long-term self-preservation runs even deeper than the desire for instant gratification. If I showed the individual a video of them dying because they attended just *one* party, they'd be (rightfully) scared and tolerate the short-term discomfort of skipping parties in exchange for getting to attend more parties in future.
I can also show you that chains *can* be centralized and insecure as a result, and that meant you couldn't build an application with the confidence that "it's never exploited or halted as a result of a chain failure". Or I can show you that, as a user, you can be censored indefinitely (have your funds frozen), and that safety attacks from not having a robust consensus layer can devalue your assets or outrightly steal them.
Obviously, you're going to be thinking twice about YOLOing into the next L1 that's offering buzzy apps, snappy UX, and fees so low you need to think up a new word for "lower than ultra-low" (whilst conveniently hiding all the subtle layers of centralization) after you become aware of these details. That's where "users don't care about decentralization" theory starts to break down pretty fast (we saw this happen when people moved from sidechains to L2s early into the rollup-centric roadmap).rekt.news/bnb-bridge-rek…
Does this mean we shouldn't care buzzy apps, snappy UX, and ultra-low fees? No. You deserve these things, and we acknowledge that high fees, awful UX, and clunky/stale apps won't have the next billion people using crypto any time soon. But we're building *up* to a scalable (rollup-centric) Ethereum after working to ensure sufficient decentralization and security at the base layer and at L2.
Solving for scalability (which means more cool apps, awemazing UX, and cheap fees) is not a cake walk, but it's certainly more tractable *after* you have solved significantly for chain decentralization and security. Solving for decentralization *after* you optimize for scalability initially is a harder and less tractable problem, however.
That's why Ethereum is choosing the harder path of solving see decentralization/security and proceeding to optimize scalability. It's making this progression steadily and pragmatically because rushing and cutting (centralization) corners to race ahead in the scaling wars is both unsustainable and a net negative for users, builders, validators, and the wider community.
So, you will have buzzy apps, snappy UX, and cheap fees at some point. You don't have to take my word for it; filter through the noise and see the people heads down and working day and night to get (rollup-centric) Ethereum to where it needs to be for the world's population to use it.
When you get the buzzy apps, snappy UX, and cheap fees, you'll enjoy these perks without worrying about what you're trading off in exchange (no "am I going to lose all my assets someday?") because Ethereum is scalable, decentralized, and secure. This is the North Star that guides the Ethereum community. Researchers, developers, and builders aren't acting randomly--in due time, what I'm saying here will take shape and cease to be sci-fi.
Now, I'm not a futurist--so maybe the future won't be as nice (e.g., dealing with awful bridging UX 10 years from now would suck). But a man is allowed to dream (Solana manlets like @0xMert_ and @pokerchessman have some (weird) dreams that something to do with "big state machines" and "synchronous, atomic composability" too). More importantly, a man is allowed to show the Ethereum vision and convince you that "we focus on decentralization because users care about decentralization".
Thanks for coming to my Sunday TED talk. Congratulations if you're still reading--I hope you're getting blue-pilled (the right kind of blue-pilling). The future of Ethereum is exciting!
Solving for scalability (which means more cool apps, awemazing UX, and cheap fees) is not a cake walk, but it's certainly more tractable *after* you have solved significantly for chain decentralization and security. Solving for decentralization *after* you optimize for scalability initially is a harder and less tractable problem, however.
That's why Ethereum is choosing the harder path of solving see decentralization/security and proceeding to optimize scalability. It's making this progression steadily and pragmatically because rushing and cutting (centralization) corners to race ahead in the scaling wars is both unsustainable and a net negative for users, builders, validators, and the wider community.
So, you will have buzzy apps, snappy UX, and cheap fees at some point. You don't have to take my word for it; filter through the noise and see the people heads down and working day and night to get (rollup-centric) Ethereum to where it needs to be for the world's population to use it.
When you get the buzzy apps, snappy UX, and cheap fees, you'll enjoy these perks without worrying about what you're trading off in exchange (no "am I going to lose all my assets someday?") because Ethereum is scalable, decentralized, and secure. This is the North Star that guides the Ethereum community. Researchers, developers, and builders aren't acting randomly--in due time, what I'm saying here will take shape and cease to be sci-fi.
Now, I'm not a futurist--so maybe the future won't be as nice (e.g., dealing with awful bridging UX 10 years from now would suck). But a man is allowed to dream (Solana manlets like @0xMert_ and @pokerchessman have some (weird) dreams that something to do with "big state machines" and "synchronous, atomic composability" too). More importantly, a man is allowed to show the Ethereum vision and convince you that "we focus on decentralization because users care about decentralization".
Thanks for coming to my Sunday TED talk. Congratulations if you're still reading--I hope you're getting blue-pilled (the right kind of blue-pilling). The future of Ethereum is exciting!
• • •
Missing some Tweet in this thread? You can try to
force a refresh
