Hacker News new | past | comments | ask | show | jobs | submit login
Ethereum Sharding FAQ (github.com)
223 points by dmmalam 12 months ago | hide | past | web | favorite | 222 comments



Something about Ethereum feels like a 19th century perpetual motion machine. "Ok, so the overbalanced wheels didn't quite work, but we'll add ball bearings and flowing water and gears and..."

In 1872, a man named John Keely demonstrated a machine that operated on what he called the "vaporic" or "etheric" force:

https://en.wikipedia.org/wiki/John_Ernst_Worrell_Keely

The same year he raised 5 million dollars, equivalent to about $100M in today's money. Two years later he demonstrated a model that was described as a "hydro-pneumatic-pulsating-vacu-engine" with "quadruple negative harmonics". No doubt he'd be a blockchain entrepreneur today.

(Edit: Someone actually maintains a big wiki dedicated to Keely's "Sympathetic Vibratory Physics" and "Etheric Elements": http://www.svpwiki.com/tiki-index.php?page=Sympathetic+Vibra... Amazing that an investment fraud from 150 years ago is still producing "research"! A sobering reminder of the power of lies.)


Perpetual motion machines are disfavored because they break a law of physics, not merely because they involve a stack of complicated mechanisms, each of which fixes a problem introduced by the mechanism below. If you've ever had a complicated medical problem, you know that very often you get drug A to counteract side effects of drug B, which was used treat an infection caused by procedure C which was a necessary precaution before performing treatment D to treat the underlying illness. Yes, sometimes such complicated mechanisms are obfuscating a doomed process, but other times engineering is just complicated. The Saturn V had a lot of interdependent parts too...


I would say the "Proof of Stake" they're planning to move towards also defies the law of physics.


Intuitively, I think "proof of stake" could violate Gödel's incompleteness theorems. These basically state that no complete system can proof its own correctness. And that's exactly what "proof of stake" tries to do: proof its correctness without external anchor of truth.

Of course, Vitalik Buterin knows that so he resorted to "weak subjectivity", basically saying that if you are not online often enough, you need to trust someone with providing you check points to decide which fork is the "true" one.


Be careful about how you interpret the word "prove" there. Are you sure that the sense of "prove" that applies to that theorem of Gödel corresponds to the sense of "prove" when you are talking about proof of stake?

The proofs of Gödel's theorems work by encoding the axioms and inference rules of the system in a way that can be dealt with within the system, and using that to encode statements like "this statement cannot be proven within the system" within the system.

I am not sure what the analogous thing would be if you were to attempt to carry the proof over. I am not even sure what it would mean to try to do that.

To prove that, there is no proof that can be carried out within proof of stake, that there is no proof of Falsum that can be carried out within proof of stake?

What would that even mean? What would it even mean for proof of stake to prove Falsum?

I am not asserting that proof of stake can work well, but I am not convinced by the line of reasoning that you lay out, with its current level of detail. Perhaps if you went into the details I would find it more convincing.


Just a quick handwavy idea: from falsity you can derive anything, so maybe in this context Falsum would be something that could be used to "prove" the validity of any transaction you want? That said, I'm not an ETH enthusiast and I don't know how you'd express this in that system or if their notion of proof is comparable to "proof" in logic.


Isn't that equivocation on the word "proof"? In the case of "proof of stake," I don't think it's referring to a formal proof of the correctness of the system. It's intended to be a consensus mechanic that the users of the blockchain consider "good enough" to trust. Of course there can always be collusion among parties in a blockchain or other attacks on the blockchain, but "proof of work" (and every "proof of x" mechanic I know of) is intended to be "good enough."


"Proof of stake" does not appear to have anything to do with "proof" in the logical sense. I really, really do not think the incompleteness theorems are at all relevant here. Apologies, but this reads just like every other nonsensical attempt to invoke the incompleteness theorems without a good understanding of what they actually say. (Like, how are you possibly going to get proof of stake to simulate arithmetic?)


Proof of stake is just another way to achieve distributed consensus. Not sure how that’s impossible or violates the laws of physics.


While Hermel certainly appears to have no idea what he's talking about, that doesn't make this reply of yours correct either. In particular, the incompleteness theorems are not physics.


I never made the assertion that Godel’s incomplete theorems are physics. My comment was to point out that if one makes the claim “X is impossible”, the burden of proof lies upon the person making the claim or assertion. Simply making an assertion or premise does not prove nor disprove it.


I do not find your explanation of the objective of PoS convincing. Maybe you are talking about their protocol that is true by construction?

Proof of Stake strategy is to make it where attacks on the blockchain are impossible in the economic sense. Proof of work strategy is making the attack infeasible by making the work required impossibly large.

Now is PoS possible? No idea. That possible fork attack seems like an impossible barrier to overcome. But we will see.


History has shown us that humans don't always act economically or rationally. So nothing is impossible.


In PoS, it is a known assumption that if the majority of nodes act rationally, then so be it. This is an issue that is out of scope.

Humans can be rational or irrational. That is how they are born. So far, we can only make tools and books to prevent irrationality.


Do you think PoS is as safe as PoW when dealing with irrationality?


Yes, there are principled ways to model these things. You can compare the byzantine fault tolerance threshold across PoW and PoS. You can compare the cost of deviating from the Nash equilibrium, and calculate how much an attacker needs to spend to deviate in return for every dollar of damage he causes.


Sorry, but you seem really confused about Godel's theorems. You're going to need to explain why you think it's related to proof of stake at all, besides the use of the word "proof".


To me the logic behind Proof of Stake is more like "so much money is bet on the answer being correct, that we'll assume it is. If the majority is wrong then feel free to permanently leave this chain and start your own". I consider it to be a superior system to Proof of Work in nearly every way.


The problem with that is that you only know that 'so much money' is being bet because you know who owns the money. But you only know who owns the money because that same money was bet earlier by people that you know owned the money because the same money was bet earlier by...

It's completely circular. It's like the Bible trying to proof itself. With no anchor to the physical world, you can't actually ever know


It's not completely circular. For example in casper-ffg if you as a full node log in every 4 months starting from ethereum's genesis block (about 3 years ago), the chain of logic you outlined terminates at the genesis allocation and PoW rewards.

I'm not saying that requiring clients to log in every 4 months is a reasonable security assumption, just showing that hand-waving it as "completely circular" is incorrect. It's more nuanced than that.


How did the node get started in the first place though? Unless you were part of the genesis block yourself, you only believe the history at all because someone told you it was the right place to start.


In full PoS in the model I described you'd need social consensus to determine the allocations in the genesis block. In PoW you need social consensus on the genesis block as well - the PoW algorithm, the initial difficulty and block reward, the difficulty adjustment algorithm, the block reward schedule


> "so much money is bet on the answer being correct, that we'll assume it is"

Good luck with that https://en.wikipedia.org/wiki/Too_big_to_fail


The one you link is entirely different. zzz


How?


Corporations is controlled by those who own the shares. 1 share 1 vote. But usually only a very very few individuals own majority of the shares. So they have all power what a "too big too fail corporation" can do.

With Ethereum PoS. It is dependent on ETH being distributed to many individuals. If only a few individuals own most of the ETH, then they will then have the power dictate what is the "truth" in the blockchain. Right now, ETH is very well distributed around the whole planet.

As you can see, one is oligarchic, the other is very democratic.

Ethereum is not a Corporation that is designed to make profits. It is non-profit, it's not even a company, it's simply a protocol implemented in Free Software.


Allow me to hyperbole this a bit. If we go back to 2008, this is like saying - "there is so much money bet on subprime loan being correct, that we'll assume it is". And it worked for sometime before it dint.


Possible. Progress will be made by sharpening that intuition into a mathematical proof or clean toy model exhibiting the failure, not by arguing from analogy.


I dunno. Proof of Stake is simply codifying an old rule of economics: The rich get richer. You prove your stake and get more Ether out of the deal. A "Dividend" if you will.

I don't think its very good for the cryptocoin community, but hey, I'm up to see the grand experiment unfold.


I feel like proof of work reduces to proof of stake because the rich get richer, there's just an intermediate step of buying miners with your dividend.


PoW is generally even less balanced than PoS, because poor people can't afford to take advantage of economies of scale with ultra-efficient mining farms. Meanwhile everyone will get the same % return in PoS.


Both are "rich get richer" schemes in the grand scheme of things, except that Bitcoin PoW started out as a completely open network which anyone could start mining from, so technically it's our own fault that we didn't realize the potential of Bitcoin in the early days and jumped onto Bitcoin. Otherwise we would be all billionaires. So I don't think it's the same. Just because you can label two things "Rich get Richer", doesn't mean they have precisely the same quality of "rich get richer scheme". Bitcoin PoW is completely capitalistic where people who mined to get their bitcoins deserve their wealth--they "worked" for their coins. So in Bitcoin it is true that "the rich get richer", but also "poor can get richer" too, as long as they started out mining early. This is not the case for PoS systems where bootstrapping PoS requires external capital, so this one is truly "Rich get Richer" in a sense that ONLY the rich can get richer.

That said, talking about this "fairness" is totally missing the point. As I mentioned, in the grand scheme of things it's all "rich get richer" and that's how capitalism works. The truly important question is how immutable an stable the ledger is, because that's what makes these blockchain projects meaningful.

If you ask anyone who has a thorough understanding of these technologies they would say PoW is superior to PoS in this sense. And Yes, this includes even the guys who are working on PoS blockchains. The only reason people bother with PoS is because PoW is having a hard time scaling and many people think maybe there may not even be an ultimate solution just with PoW.


... there's the other small detail of the appalling electricity consumption of PoW. Blockchains with PoS are more appealing to people who care about the rest of the world.


Congrats you've just repeated myself. Look for the part where I said PoW doesn't scale.

You probably didn't even read my comment. Because it wasn't about whether PoW or PoS is more efficient.


That’s a rude way to say “thanks for clarifying, I agree with you.”

“doesn’t scale” usually refers to technical difficulty with throughput. If you meant that Bitcoin et al cause massive environmental damage, you sure didn’t say it clearly.

I suggest that you would get a lot of mileage out of improving your writing skills first before trying to flame people. As it is, you’re just playing yourself.


PoW is theoretically better in this regard.

I mean, in theory the market should push the profit down to a minimum, forcing the miners to spend their revenues in order to keep mining.


It is however extremely more inclusive. Everyone in the world right now has very good chance to join. Unlike traditional capitalism where it can do lots of things to keep others from joining, keep many dependent on their salaries just to survive.

It won't be perfect. But it is definitely tons better than the current capitalism.

Another very important feature here is the ability to easily fork. So if we need to fix something, it can easily be done (no need to wait for an organized citizen revolution)(although this is probably better attributed to the idea of Free Software).


No better than the stock market in this regards. Except the stock market explicitly hands out profits from the biggest and riches companies, and our tax system is designed to give more of those profits to poor people.

A low-income person getting into Ethereum today has no hope to ever get as much ethereum as anyone else in the early adopters.

Hell, a low-income person was too busy trying to buy food today and/or pay off credit card debt to even think about investing into Ethereum, the Stock Market, Bonds, or other securities.


Pretty sure getting into Ethereum is easier and fairer than getting into Stock Market.

If you wanted to be an early investor of facebook and you are a pleb, then you're out of luck. And you should have a big capital to join.

If you wanted to be an early adopter of Ethereum, then you only need an internet connection and a computer. You invest put practically any amount, so any pleb can join.

There's also a law that requires you to be big capitalist before you can do early big investments right?

Pretty sure this is better than stock market.


> Pretty sure getting into Ethereum is easier and fairer than getting into Stock Market.

Robinhood offers free trades on anything in the Stock Market. Literally free, you can't beat that.

> There's also a law that requires you to be big capitalist before you can do early big investments right?

That's not the stock market. That's venture capitalism. The stock market is literally open to everybody with an internet connection from a variety of well trusted sources (RobinHood, E-trade, TD Ameritrade, Vanguard, etc. etc.) and a variety of pricing schemes. ($0 for Robinhood, although it seems like their execution is not as good. You get what you pay for afterall)

Furthermore: the stock market is regulated. The National Best Bid and Offer (NBBO) system in the USA ensures that everyone in the USA pays relatively consistent prices, or at least more-consistent than BTC / ETH. Unlike ETH or BTC, no one is going to steal your money Mt. Gox style as the SEC is a strict watchdog.

You get no assurances with Eth or BTC. Coinbase has different prices than Bitfinex. And some of these exchanges look like they're running scams (ie: Bitfinex's support of Tethercoin).

I think more "plebs" have been burned by Mt. Gox and other such failures in the cryptocoin marketplace. Its a hugely unregulated, and highly risky market. Cryptocoins are just not mature enough for mainstream speculation yet.


Ok, I stand corrected. It was Venture capitalism I had in mind.

I wanna note though, regarding crypto exchanges. One does not need to use them to invest. Use your own wallet. Use decentralized exchange if you don't wanna risk using centralized exchange.


In what way?


You have to at least give him credit, he didn't use a car metaphor.


> Something about Ethereum feels like a 19th century perpetual motion machine.

Funny you'd say that:

"Vitalik's project immediately before Ethereum is that he was collecting investments from people to fund building a computer program to solve NP-complete problems in polynomial time, supposedly by simulating a quantum computer."

http://archive.is/8wHvq

"Buterin later said he had “greatly overestimated” the likelihood of the team breaking mathematics"

https://davidgerard.co.uk/blockchain/buterins-quantum-quest/

More: http://bitcoinerrorlog.com/2016/08/vitaliks-quantum-quest/


Simulating a quantum computer on a regular computer to speed up np-complete problems is insanity. Trying to break mathematics is insanity.

I tried to verify your sources, it seems to go back to Gregory Maxwell and an anonymous source, BitcoinErrorLog. David Gerard cites BitcoinErrorLog. zero proof.


https://twitter.com/peterktodd/status/906189639802859520

Jorge Timón was pitched it

> Was he soliciting investment though? > Jorge Timón @timoncc Yes, he and another guy were claiming the classical computers simulating quantum would be a mining revolution.


http://archive.is/bXZaW

interesting, but that sounds quiet differently from the claims from maxwell and your original post.


Did you expect him to say, "Yes you caught me guys!" - instead he blames it on "a team" and it only had 5% chance of working, etc.

He has never denied trying to get money for a quantum project. If rumors were floating around that I had I'd instantly deny it 100%


so vitalik says he just had a short term connection to this team before he grew suspicious. if he really was the head behind a scam, why did noone sue him? why is there no evidence, like former employees, payments, company registration?


> why did noone sue him?

perhaps because not a single person was dumb enough to fall for it?

How about this- for 4 years people have been talking about this, not once has he simply replied "I never tried to raise money to build a quantum bitcoin miner"


well, i guess he gave a talk about quantum computing and bitcoin, you could see that as an ad for that company. this looks to me like a motte-and-bailey argument. if he gave a talk, grew suspicious of their plans, quit supporting them, then there is no wrongdoing here, no proof that he ran a scam company.


i'm honestly trying to verify this. i'm quiet skeptical about the character of some developers of bitcoin, i dont know about peter todd. bitcoinerrorlog is linking to this video of him talking about quantum computers as a proof. this is the closest i have come to hard evidence of this extreme claim. i have listened to the first half of the video, just him talking about quantum computers. Very impressive talk for a 19 year old.

https://youtu.be/DkUpZkeqhF4?t=16m46s


How about Vitalik himself

http://archive.is/bXZaW


Fwiw bitcoin error log is a real human being who attends conferences and has met many people in meatspace.


He went to conferences with that pitch trying to raise interest and capital. A lot of people who were active back then hard the pitch in person. It's not exactly a secret and no reason trying to pin this on anonymous sources.


if you read my other comments to nope96 in this branch, you will be able to understand my position better. i even linked the video bitcoinerrorlog cites as proof, which is probably the talk you are referring to. none of this proves he was the head of this company, it does not prove that he knew their qc was never going to work. it does not show him lying, in the video he talks about qc and the consequences for bitcoin.


Vitalik admitted it himself. I linked the ref in my piece.


i have read that, maybe you should read it again.

http://archive.is/bXZaW


All sufficiently successful entrepreneurial “geniuses” are eventually shown to have been lucky for the most part.

  - Proverb


Funny that you mention the 19th century, because another example of a system that required an ever-increasing series of hacks to solve major flaws is the internal combustion engine. Whether the problem that Ethereum solves is anywhere near as valuable as the problems solved by ICE remains to be seen, but this just seems like the engineering process to me, especially for a young technology.


Right! There was widespread public enthusiasm about engines in the latter half of the 19th century. Both John Keely and Karl Benz created engine startups, among hundreds of others around the world. Keely ripped off gullible investors for decades with faked demos, while Benz's company is still going strong.

But if we're going to make the analogy to the present day, it's important to frame it right. Blockchains are just a subset of distributed databases, and the lasting "Benzes" of the field might be projects like FoundationDB rather than blockchain-anything.


Internal combustion engines didn't have research papers explaining why they were impossible.


Internal combustions work, really well, today, and are not theoretically impossible. Not really the same.


Well, you haven't proven Ethereum's theoretical impossibility either. 'I think this might be a scam' is not a proof, no matter how many heuristics are on your side.


https://download.wpsoftware.net/bitcoin/pos.pdf

This paper does a pretty good job of laying out a formal, mathematical proof that pos doesn't work


They work really well thanks to the invention of the spark plug, the carburetor, the supercharger, and decades of iteration. For many years, they were objectively terrible.

Ethereum is three years old. ICE is 200+.


Somehow I think that at 3 year anniversary of ICE people had some actual real world use cases for ICE where ICE was still somehow better than steam engine and animals, even if ICE itself was objectively terrible?


Ethereum is objectively better in some cases. What if I want to gamble somewhere, but I don't trust websites to be honest about their odds? With Ethereum I could check the source code of the contract and be confident that it's legitimate.


Not only can you verify the code, you can verify that the code and the output wasn't modified during execution by some external actor.


A 1910s ICE might have been terrible compared to a modern computer-driven engine built on one or the other dozen million man-hours, but back in the day it was still a very useful thing that powered machines and vehicles and even had broad military applications.


ICE didn't work really well in 1910.

Ethereum works today as well.

I think you missed the point of the analogy.


I'm sorry, but I feel this is really lazy and nonconstructive criticism. Perpertual motion machines violate a law of physics, and the bells and whistles attached to it were at best misguided attempts that didn't address the fundamental issue, or at worst attempts to defraud investors. I can't tell what you are claiming here; is it that

1. Sharding is fundamentally impossible; in any public blockchain, all computers must validate all transactions, and the scalability trilemma cannot be avoided, or

2. Sharding is possible, but the solutions outlined in the FAQ are unnecessarily complicated; a much simpler solution exists

You haven't provided arguments for either, and I can't even tell which argument you're making.

On a higher level, I could agree with the general sentiment that complicated solutions tend to be wrong or at least very suboptimal, but only as a heuristic. It's a heuristic that doesn't even hold very well in this space; in the space of PoW fork-choice rules, bitcoin-style longest-chain is simpler than Somplinsky's GHOST rule (https://eprint.iacr.org/2013/881.pdf), which is in turn simpler than ethereum's current "modified GHOST", but in each case the added complications really were necessary and lead to better properties.


More explicitly, my position is this: Ethereum's claims of an upcoming blockchain-powered "Web 3.0" revolution are so overblown that they're indistinguishable from perpetual motion peddlers.

The Ethereum designers don't want to recant their outlandish promises, so they keep proposing increasingly complex bells-and-whistles fixes that sound impressive in a 10,000 word FAQ but won't bring the "global supercomputer" substantially any closer to reality.


Great, if frustrating reading there. I love the linked news article interviews with Lieutenant E.L. Zalinski.[1][2]

> Lieut, Zalenski's mirth interrupted him at this point. "Now," he resumed," when I started for Philadelphia I was careful to take with me a 10,000-pound gauge. That was my little equipment. This, as I suppose you know, or if you don't must take my word for it, is a standard instrument. Well, as soon as the Keely people saw this gauge they said to me, 'You had better not show it to Keely. It would only anger him!' Much I cared about angering him! I wanted justice. One of his partisans, however, informed Mr. Keely that I had brought this gauge with me, and the great man replied that he would not use it. In the course of his experiments he said he had got 50,000 pounds pressure. But—observe the but—he then remarked that he had broken all the pressure gauges he had. I then quietly rejoined that I had a pressure gauge which was only capable of registering 10,000 pounds. Said I, 'I would like to have you put it on, and break it for me.' Mr. Keely made no immediate reply to this. Shortly afterward, however, he remarked, with an air of feeble assertion, 'I do not believe in pressure gauges, anyhow.'"

> The subject appeared to be excruciatingly funny to Liet. Zalenski.

[1] https://timesmachine.nytimes.com/timesmachine/1884/09/24/106...

[2] https://timesmachine.nytimes.com/timesmachine/1884/11/16/106...


Those are just brilliantly written, thank you. What a contrast in journalistic style to today!

"And the energetic gentleman laughed in the same strain in which the cat was heard when she feelingly exclaimed, 'Hi, Hi! I have eaten the canary!'"

"The 'etherealized vaporic force' which was so liberally poured into the ears of a large and gullible public..."

I can't get enough. It's wonderful.


Reminds me of that guy who pretends to be Satoshi Nakamoto: https://arstechnica.com/information-technology/2016/08/craig...


As someone deeply involved in this community, I sort of agree about the "Rube Goldberg" feel of his ideas, but Vitalik et al have delivered pretty reliably: I think both ethereum and casper are pretty safe bets at this point (yes, I'll likely get some comments telling me I'm wrong and how horrible ethereum and casper are)

The sharding stuff, however, is still iffy at the moment (I've studied it pretty closely) ...but if vitalik is confident he can make it work, I'll bet on v every time.


Ethereum is, so far, missing a key element for me to take it really seriously: Formal Methods.

There are some small efforts, but I'd rather have formal methods from the beginning, not as an afterthought.

The problem is that market forces favor releasing early rather than releasing correct things. It's the MIT vs NJ paradox again, or worse is better. But this time the potential risks are much bigger.


well, that would be tooling though. you wouldn’t say that x86 is missing formal methods therefor you shouldn’t use it right...

solidity, etc sure maybe... but at that point, releasing early is probably a good thing so that you can get a whole community working on things and getting ideas and collaboration from a large group of people

*edit: unless you mean like geth, or some minimal reference client should be formally proven


Well, a lot of x86 actually does get proven using formal methods. Clearly not all of it, but we've realized that was a bad plan, and that we should go back to formally verifying all of our hardware.


Well it was a ton of work to get ethereum out the door... Adding a couple more years to formally prove everything first sounds like the perfect way to be stuck in development limbo forever. As you say, formal methods are being applied after the fact to some degree at least.


I think it is a common misconception that formal methods will lead to an extremely slow development cycle. That is only the case if you go for the best guarantee of correctness: Theorem Proving.

There are more agile options. For example, model checking all protocols is pretty easy in comparison to theorem proving. Still a significant effort, but perhaps an order of magnitude easier.


Uh, I think that many cryptocurrency experts would take strong opposition to the idea that ethereum has delivered reliably.

Just because a bridge is still standing today doesn't mean it's safe. Ethereum has been almost religious in violating best practice blockchain design.


Yeah, I looked at the original Ethereum spec and thought this thing isn't going to scale. Boy, did I miss out on making a lot of money. I still don't know what practical use people actually use Ethereum for besides as a currency and ICO scams.


One use case that's already operational and doesn't need additional transactions - distributed domain resolution

The entire .eth registar runs in a smart contract. You put your ETH in to claim the domain, others also bid during the bidding period, the winners ETH is locked away and returned when they release the domain, the rest get refunded.

Considering DNS is continually politicized and still subject to points of failure (the CAs), distributed ENS is a far better design IMO


There's a decentralized Twitter clone that runs on the Ethereum blockchain. https://peepeth.com/welcome


Haha, “Vitalik’s project immediately before Ethereum is that he was collecting investments from people to fund building a computer program to solve NP-complete problems in polynomial time, supposedly by simulating a quantum computer."


He was like 16 years old when he was working on that thing and has a solid track record since then... let's cut him a break.


He tried and failed to scam people when he was 16, but now has a successful scam so we should cut him a break?

"solid track record" meaning cloning bitcoin and giving yourself and friends a bunch of free tokens worth $7,764,000,000:

"12 Million Ether were created to the development fund, most of it going to early contributors and developers"



What solid track record? I see just hype and broken architecture.


"solid track record"

Can you name one thing he built/invented that does anything? Beyond collecting bitcoin.


Speaking of which, Ethereum is not really novel either. Bitcoin includes its own VM which has opcodes that allow smart contracts too... just most of them are disabled because the reference implementation had bugs at the time Satoshi abandoned the project, and the development process is currently so calcified and politicized that development has essentially ground to a halt.

The concept of memory-hardness is not really novel either, see: bcrypt/scrypt.


Vitalk was originally a bitcoin developer who proposed extending the op codes to make a more general purpose computer, and only started Ethereum after he encountered resistance to the idea. I don’t think anyone was claiming bitcoin wasn’t prior work.


Can you cite any evidence at all that he actually proposed new op codes? I've heard this claim many times, but never seen any evidence to back it up.


What opcodes does Ethereum support that Bitcoin does not support, or could not be implemented using Bitcoin's opcodes?

You've got everything that looks essential to me: add/sub, mul/div, logical and/or, bitwise, string, crypto, etc.

https://en.bitcoin.it/wiki/Script#Opcodes

No disagreement that the Bitcoin development process is broken as fuck though.


Bitcoin specifically avoids loops and recursion. The codes you included allow basic contracts, but bitcoin scripts are guaranteed to halt. Its a conscious decision, and I'm not sure its a bad choice for the project.


here is vitaliks version of the events:

http://archive.is/bXZaW


I agree, I don't think there's any clear evidence that they will eventually get Ethereum to work at scale. At one company I was with, there was an experimental project that only had "just one more thing to fix before it will work" for ten years, before being eventually abandoned.


The name isn't a coincidence. I kind of agree with you, there's a whole lot of words and invented terminology but not much is being said... however solving the problems DLT currently has is (hopefully) doable even if the "well let's add XYZ to the ABC!" approach fails.


Yep. I've considered many times that Ethereum went down one path of evolution from Bitcoin, which is get more and more complicated. The problems with certain subsystems (the scripting language for example) have been discussed at length on HN, knowing how hard it is to modify complex systems that only get more complicated, I doubt they'll ever fix these issues, they'll be battling them indefinitely. That's a real problem considering you can never really trust Ethereum and the entire thing is predicated on trust.

And what does all this complexity buy? Cryptokitties. We sure this is a worthy tradeoff? Would they Eth guys even be able to recognize this considering what's invested personally and monetarily? There's plenty of suckers, so I'm sure the "next year" kicking the can down the road cycle can go on as long as it's needed, but I'm interested to see if we ever get something of value out of it.

The other evolutionary route interests me, but I'm too busy, lazy, ignorant, and disturbed by the entire space to test it out. Maybe some day.


I guess you aren't familiar with MakerDAO and DAI.

etheraffle.com also seems like a relatively good use case for a ÐApp.

In fact there are thousands of good use cases.

I don't think you're familiar enough with the space to be providing meaningful criticism. Your reductio ad absurdum to "just cryptokitties" is fairly telling.


I've been observing the space since before Eth launched. If you believe that raffles are an example of useful systems, then we'll have to disagree.

In fact there are thousands of good use cases.

If there are, then we'll surely see them. You're providing more of what I'm talking about - it's always next year, always coming, but the mainstream adoption is nascent.


Any comment on MakerDAO + DAI[1]? I think it's an incredible project / feat of engineering and would be interested to hear how you think it's just a toy that doesn't serve a good purpose.

Whether or not you "believe" raffles/lotteries to be a useful system does not change the fact that it was a system with a $280BN market cap in 2015. Lotteries on a blockchain with smart contracts solve three problems with traditional lotteries: transparency (provable fairness), easy and secure participation, easy and secure disbursement.

  We'll surely see them.
Yes! You can see them right now. There are thousands of DApps, at different stages of development[2].

1: https://makerdao.com/

2: https://www.stateofthedapps.com/


It's another instrument (like lotteries) whose main value is to make money for the founders. That has a use, yes, but if you're outside of that system I'm not sure the mainstream value is that great.

I can tell you that the team is the least diverse (not just race/gender but in age/possible life experience) I've ever seen, which doesn't bode well. How different can their perspectives be?

https://makerdao.com/team/

The vast majority of those apps are cheap moneymaking schemes. Again, they have a place, but I'll very much take a wait and see approach, which I believe is realistic and more than warranted.


I'm not seeing why race/gender/age are relevant. These are all superficially diverse things that don't actually say anything about diversity of perspectives, unless you believe that people who belong to the same group also have similar thoughts.


Dai is a stable cryptocurrency, not a speculative money making scheme, since it's purpose is to be stable. A decentralized stablecoin is an enormous use case.


With a name like "MakerDao", one would think it would have something to do with the Maker movement, in that it'd be connected to fabrication tools somehow.


While I suspect that in the long run blockchains will absolutely end up finding some niches where they're persistently used, I do agree that a lot of the broader claims remind me a lot of "next year will be the year of linux on the desktop".


Fast forward to 2018, and we have Windows Subsystem for Linux, and a very popular BSD in silver aluminum cases made by a fruit company, dominate the desktop market. You could say that the year of linux on desktop is today (in some perverse modified form, but nonetheless).


That's actually a really good point (especially if one includes Android) - but "20 years later than we hoped, in a completely different form than expected" is significantly what I was trying to handwave at.


The web was nearly useless and mainly business card style web pages from 94-98 even though the building blocks were available.


That's just not true. The 1996 web already had news, porn, and forums — the same things people use it for today.

Those things were a bit slow to access over dial-up (porn in particular), but millions of people were using the web daily by the time Netscape IPO'ed.

There's no similar moment in sight for Ethereum's hypothetical "Web 3.0".


I'm claiming that the realized potential was maybe 1/100th of what we have today. Perhaps I should have left 98 out, but the idea stands. And that's after the component parts had been around for years: https://www.livescience.com/20727-internet-history.html


1. The web was already used for all sorts of things in that period. Geocities was a wonderful open world; I have fond memories of Simpsons and X-Files fan webrings. Not to mention services like ebay, Amazon, or Craigslist arose during that period.

2. Even if we accept your statement as true, the world wide web became generally available in 1991. Blockchain technology has now been around for a decade. The web certainly showed its practical usefulness in its first 10 years, yet blockchain remains largely in the realm of hype.


Yea but where was the web in 1981 or Arpanet in 1969. Bitcoin was just an idea in 2008, didn't gain any real significance until 2013 where it really started to enter mainstream consciousness. If we want to compare it to the web we're still far away from 1991 in terms of tech.


"Blockchain technology" is a misnomer because different projects require different technologies and concepts.

Ethereum was the first blockchain with smart contracts and it only went live in July 2015...


And was immediately and irredeemably invalidated in May 2016 with the DAO fiasco, yes.


Ethereum Classic lives on, feel free to support it with money/code if you feel so strongly about immutability at all costs


EXACTLY this.

I have yet to see any good use case for smart contracts, except for, of course, launching your own speculative asset.

I just want to see a real use case that has nothing to do with launching more tokens.


Or the epicycles of the heliocentric era, where the focus then as now was on futile ever finer retweaking of conclusions to fit unexamined and faulty premises.


One theory I have on the hate towards Ethereum (and cryptocurrencies / blockchain in general) on HN is that it's a radically different way of solving problems, something that is alien to software engineers who have put decades of time becoming experts at solving problems a certain way. So rather than invest more than 15 minutes learning about it, they say "Blockchain is just a distributed database and it's useless!" or attack it from a variety of minimum-thought / minimum-effort angles.

There is genuine innovation in this space and it is going to impact a lot of areas. Decentralization, immutability, protocols monetized with market forces, fee-less payments that cannot be prevented, to name but a few interesting areas.

And of course there are numerous downsides. Difficulty of keeping keys safe, bugs that cost millions of dollars, issues with scaling.

But these are problems that people are solving right now. As a technologist, you should be looking at the possibilities! Now is the time to start a new company or project, not 5 or 10 years from now when the tech has matured.


I think a lot of the hate is a reaction to the baseless hype from people just trying to get rich from cryptocurrency. There are tons of altcoins and ICOs being made and advertised constantly that are nothing but hot air. There are people making a lot of money (or thinking they can) from buying into altcoins (hyped up by nonsense marketing pages), which they then try to shill to others. People are often being incentivized by rising prices to act this way!

Outsiders start to learn that "blockchain" = "get-rich-quick nonsense", and miss the interesting technical discussions (which often come from the Bitcoin and Ethereum spaces).


Yes, this is definitely a big part of it. HN used to be super into cryptos, when it was more about the technology (pre-2017, or even pre-2013). That being said, the pendulum has pretty clearly swung too far the other way here.

The pretty obvious reality is that smart contracts and cryptocurrencies are truly interesting and novel technology. Does that mean they're good investments? No. Are they going to take over the world? Maybe, maybe not. But they're definitely 'cool', in the technological/hacker sense. They're cool in the way the internet was cool - they enable a fundamentally novel form of human interaction. It remains to be seen if that form of interaction will be economically meaningful. But the novelty and possibility of it is definitely interesting, and anyone who says otherwise either doesn't get it, or has been jaded by the scams. An understandable response to all the greed and general chicanery of course, but the reality is still that this stuff is actually pretty cool tech, even if you never use it to buy a latte.


>>I think a lot of the hate is a reaction to the baseless hype from people just trying to get rich from cryptocurrency.

This is it exactly. People are naturally suspicious of anything that is widely promoted for the profit of the promoters.


I don’t think you’re wrong or anything, but I do find it frustrating that HackerNews seems to be full of people dismissive of the technology and from what I’ve gathered relatively devoid of the baseless shilling. Other communities, not so much.


Well, baseless naysayers are just as bad as baseless hypers.


> Now is the time to start a new company or project, not 5 or 10 years from now when the tech has matured.

Actually, I'd rather start the new company 5 or 10 years from now when the tech has matured. It's the same reason the company I'm currently on relies primarily on boring PHP stuff and the more boring parts of Docker, not the newer tech that I'd rather be using. The time that could be spent chasing stuff that might be fads or might be an advantage is better spent directly working with customers to better understand their problems and exploring directly-business-relevant parts of the solution space.

That and the blockchain really is "just" a distributed database. It's potentially very useful, because of the way it tackles the trust problems, but if you're not sure whether you need it or not, then you probably don'.


As a technologist I look at vaporware and call it vaporware.

I look at get-rich-quick schemes and call them get-rich-quick schemes.

I look at how blockchain don't solve a single problem that hasn't already been solved more efficiently on much grander scales, and I call them out as such.

Your words, and images of grandeur, and promises of great things to come are just that: words, and images, and promises, and I call them as such.

Blockchain is not "the next Internet that haters cannot see", as the overused saying goes. So far it's nothing but the next Tomagotchi or Pets.com, and nothing that's happening in blockchain can prove it otherwise. And so, we call it as such.


> I look at how blockchain don't solve a single problem that hasn't already been solved more efficiently on much grander scales, and I call them out as such.

This position is the most confusing of all the blockchain FUD out there. It's easy to show that it's false: today, people are able to purchase proscribed plants and medicines in a way that completely subverts the efforts of state to prevent that.

Now, I'm not saying that's sufficient to prove the kind of meaning you're looking for, but doesn't it give you the slightest bit of pause? Doesn't it make you say, "Wow - that's something people have been trying to do since the garden of eden and nothing has ever done it so successfully. I wonder what else this thing can do?"

I think that for most open-minded observers, something like this thought process is what keeps our attention.

I think that undermining drug prohibition is a wonderful thing and an obvious net good for the world. But I also think that it's a sign that there are more mundane / lawful-good elements that will arise in time.

Porn has largely driven the internet forward at various critical junctures. But if we had stopped and said, "oh, is this thing just about porn? How useless," then we'd be in a much worse place.

Give it some time and an open mind.


>people are able to purchase proscribed plants and medicines in a way that completely subverts the efforts of state to prevent that.

No it doesn't. This is the kind of absurd nonsense that earns blockchain proponents this level of scorn.

People could buy drugs before. Sure they can use cryptocurrency instead of cash, but the old problems remain: getting the product from the buyer to the seller. The state can, and sometimes does, prevent that because until teleportation is invented getting the drugs you bought online into your hands opens you to risk. Buying drugs is certainly easier, but it doesn't help to talk in these ridiculous terms. I bought pot in college just fine. If I buy drugs and have it shipped to me I'm at the same risk whether I used cash or Bitcoin.

>"Wow - that's something people have been trying to do since the garden of eden and nothing has ever done it so successfully. I wonder what else this thing can do?"

You've got this backwards. It's been 10 years. People have been asking all that time "so what else can you do", and the answer so far is...nothing, really. This despite the relentless, utopian hype all during that time.

If proponents want people to keep an open mind and not be so cynical, maybe they shouldn't have spent years promising that blockchain technology is the digital coming of Jesus that will completely change everything in every way and usher in an endless golden age of peace and prosperity.


How do you send cash to your drug dealer on the internet, instantly?

I agree with you that you still have to get the product to the person, and blockchain doesn't solve this. But it doesn't claim to.

What cryptocurrencies do is solve the payment part of it, so that you can send money to someone without your bank finding out and freezing your account.

If I send money to coinbase, buy Bitcoin, and then convert that to Monero, the only thing my bank knows is that I sent money to coinbase. And coinbase only knows that I sent money to a different exchange.


> How do you send cash to your drug dealer on the internet, instantly?

By instantly you mean "up to 200 hours"? https://blockchain.info/charts/avg-confirmation-time (at the time of writing this comment the spoke on the chat was at 12000 minutes).

> the only thing my bank knows is that I sent money to coinbase. And coinbase only knows that I sent money to a different exchange.

Money laundering is as old an art as the banking services themselves. I guess blockchain made it easier then? True innovation.


By instantly I mean, on average around 10 minutes, if you pay an appropriate fee. This is miles ahead of credit cards, for example, as that takes days to clear. Yes they do, as the transaction isn't confirmed until days later.

Obviously, if you pay too low of a fee, it could take a long time.

> Money laundering is as old an art as the banking services themselves. I guess blockchain made it easier then? True innovation.

Indeed it does. The whole point of bitcoin was to enable censorship resistant financial transactions.

And it turns out, that the people who most want to have access to censorship resistant transactions, are the people who are CURRENTLY being censored by the existing financial system.

I have no idea how I'd be able to quickly send cash to someone on the other side of the world, in a way that is censorship resistant. Bitcoin enables this, though.

Obviously, though, if you don't care about fighting financial censorship, then you wouldn't see the usecase for a toll that helps you make censorship resistant transactions.

It is perfectly fine for you to not care about subverting financial censorship. But you should be aware that there are other people with different values, who care a whole lot about this cause.


[flagged]


Medical patients are NOT the ones using Silk Road to buy pot. You're just demonstrating a level of out of touch that earns the Crypto sphere it's scorn.


[flagged]


And as I already said before, this has a much better, more effective solution than blockchain.

Proper healthcare system and legalisation of cannabis for example.


Do you realize how tone-deaf this sounds?

Are people seriously supposed to what, wait until all the political misconfiguration of the world is solved to have relief?

Darknets and cryptocurrencies are disrupting these prohibitions today. How is this not cause for celebration?


Because they're not. They're not doing anything that wasn't already been done before.


Wow. I write: blockchain doesn't solve a single problem that hasn't been solved.

You: But you can buy drugs! Be open-minded!

So yes, ten years in, nobody has come up with a use for blockchain: https://hackernoon.com/ten-years-in-nobody-has-come-up-with-...

Oh, and yes, porn has advanced the internet. Please show me how blockchain "has driven the internet forward at critical junctures". No, buying drugs isn't it.


I really don't understand what you're talking about.

Firstly: yeah, can you squarely address the simple fact that blockchain tech has helped people acquire prohibited plants and compounds? People who previously didn't have access? That alone is enough to prove that the vehicle is not in neutral - it hasn't achieved literally nothing.

And second, you seem to have entirely whooshed over the point about porn and the internet - and to be clear, I didn't make this point up, but you seem not to have ever heard of it? Or something?

It's simple: porn pushed the internet forward. That doens't prove that the internet is worthless (or that it is worth only as much as the porn it can carry). Drug sales have pushed blockchain tech forward. That doesn't prove that blockchain tech is only as interesting as the drug commerce it can facilitate.

I'm just saying: be patient. Ten years isn't a very long time in this context. Well-intentioned, thoughtful people are working on it. The fact that it can seemingly overcome a long-time and entrenched evil in the world says that there's something there. Let's work out what it is.

Why so quick to judge?


First. Any other use cases than buying illegal drugs?

Second. What whooshed is your understanding of what I wrote. I know that porn had advanced the internet. What I asked is for you to provide examples how blockchain has advanced the internet the same way porn has.

Ten years is eternity in IT.

For a much better overview I refer you to the aforementioned article, https://hackernoon.com/ten-years-in-nobody-has-come-up-with-... and its second part, "Blockchain is not only crappy technology but a bad vision for the future", https://medium.com/@kaistinchcombe/decentralized-and-trustle...

The second part talks about drugs as well.


I have already read both of those. Here is my long-form response, posted on HN after the second link was posted:

https://news.ycombinator.com/item?id=16788447

You are seriously dodging the conversation here, but I'll try one more time, just to make this clear:

* First: Imagine my answer, for the moment, is "no." So what? Disrupting drug prohibition is already not nothing.

* Second: Yeah, you whoosed. I didn't say that blockchain tech had advanced the internet (like porn has), but that drug commerce has advanced blockchain tech. See? In both cases, the resulting boost isn't limited to porn or drugs.

> What I asked is for you to provide examples how blockchain has advanced the internet the same way porn has.

I didn't make that claim. I don't think that it's clear that blockchain tech is even cognizable as part of what we today call "the internet." So I'm not making the statement you're asking me to defend.

What I am saying is that it's not reasonable to conclude "Well, it doesn't do anything but facilitate drug commerce" any more than you can sensibly conclude, about the early days of the internet, "It doesn't do anything other than facilitate porn."

Get it?


> So far it's nothing but the next Tomagotchi or Pets.com

At least with Pets.com you could order a 40 pound bag of dog food and get it delivered to your door. It may have been a bad business model but it still tried to solve a real problem.

Blockchain doesn't even solve real problems. It is just a technology in search of a problem. So far, the only "problem" it solves is providing a world-class platform for fraudsters and scammers to conduct business on top of.

Ethereum, in particular, is an excellent platform to spin up an ICO, take a bunch of suckers money, exit scam, and then go move to a country with no extradition treaty with the US...


Problem: Want to part many fools from many millions of dollars.

Solution: Blockchain!

I think that’s the frame rails this whole messy thing is built on, loveingly maintained by scammers and some True Believers.


Thanks, you said it way better than I could.

Blockchain is a world class way to solve the problem of how to part fools and their money.


> I look at how blockchain don't solve a single problem that hasn't already been solved more efficiently on much grander scales, and I call them out as such.

What about the problem of running an immutable log using compute power around the world without paying for the costs? That's what santoshi accomplished. He stopped mining many years ago and it's been committing logs every 10 minutes for years.


You are proving the OPs point and attacking a strawman.


People said that about motor cars, about touchscreen phones, and personal computers, the internet itself... any new technology will bring with it controversy, naysayers, early adopters, opportunities and pain.

Will Blockchain technology as we know it eventually go on and become something we all use / interact with? Nobody knows... but a technologist doesn't write off new technology just because it's not in use by everyone on the planet after less than a decade.


A lot of the hate from me and others is precisely the opposite - it doesn't solve problems. It doesn't bring novel solutions to the table. In fact anything at all blockchain related seems to be putting the cart before the horse. Every time, people start from blockchain and look for somewhere to crobar it in.

In the end it's just another data structure. It may have its niche, but as yet it seems to be hype over substance.


I see a lot of this as-well, and it goes with the atmosphere of our industry. Remember "cloud" was a solution looking for a problem also: server colocation had been around for 30 years prior...

We are in the adoption chasm of blockchain architecture right now; if we make it to the other side we'll have a new wave of innovation in the distributed software space.


> Remember "cloud" was a solution looking for a problem also: server colocation had been around for 30 years prior

True, although "you get root on it", "it takes 2 minutes to set up", and "it costs $10/mo" were properties that never really lived together before "the cloud". The value prop for cloud VMs was immediately obvious for a lot of people. And, at least in the crowd I roll with, managed systems like Google App Engine had very obvious utility out of the gate too (I can just push Python code to it and it'll manage all the servers for me? Woah!)


My "hate" for cryptos is not because it is a different way to solve a problem. It is because cryptos are trying to disrupt financial industry by adding technology to medieval monetary policies that we have long history showing that they are bad policies. Putting lipstock to a pig won't make the pig fly...


Inflationary vs deflationary only really matters when you are talking about world wide level scales for things.

But even so, there are lots of modern assets in the world that are deflationary. Stocks is a great example of a "deflationary" good. Or commodities.


so if you can’t fix the problem then you shouldn’t even try to make it better? should we roll back internet banking, eftpos, etc?


If you can't fix the problem you should not try to make it worse. Internet banking etc. are making things better. I do not know how bitcoin is supposed to make anything better for a non-criminal person.


I think the parent meant deflationary economics vs inflationary of today


That would be an irrational reason to hate cryptocurrencies, because they can be either inflationary or deflationary.


It's not about inflationary vs deflationary, but adjustable monetary supply based on need vs. predefined monetary supply.


So what is your solution? Marxism? Communism? Anarchy?


Well, first issue in current monetary policy that would need fixing is removing zero lower bound from interest rates to allow heavily negative rates when needed to manage aggregate demand more efficiently. (I understand that most people do not agree to this statement, but yes, to me it looks like aggregate demand needs to be managed in economy. And I do not know that many other tools than interest rate for that.)


Can we please stop labelling doubt and maybe cynicism as hate? It is nonsensical and doesn't add to the discussion. It rather makes people even more defensive and argue their position even harder.


I would say these are the solutions people are producing right now and the problems they are solving are still unknown.


I personally have animosity toward a lot of them because they've spiked the price of graphics cards to crazy levels.


This is a typical condescending blockchain enthusiast line. Stop implying that the critics don't understand how things work, it's insulting, intellectually lazy and false. When the blockchain is actually useful enthusiasts won't have to beg everyone to give it a chance.


I've been actively researching use cases and properties of cryptocurrencies (more than 15 minutes) and have come to the opposite conclusion: none of the hype is based in fact. For instance, let's go through your claims one by one.

> Decentralization,

There is surmounting evidence that this is not true and that Bitcoin and Ethereum are actually quite centralized and would benefit from being more centralized [0][1][2][3][4].

[0] https://arxiv.org/abs/1801.03998

[1] https://freedom-to-tinker.com/2015/07/28/analyzing-the-2013-...

[2] https://arewedecentralizedyet.com

[3] https://fc18.ifca.ai/bitcoin/papers/bitcoin18-final13.pdf

[4] https://www.tse-fr.eu/sites/default/files/TSE/documents/doc/...

> immutability,

I'm not sure if you've heard of theDAO fork [5], but basically transactions on Ethereum are not final. They can be reversed if a select group of powerful individuals decide to do so. Maybe it's "immutable" in the sense that the history can't be changed, but it's not "immutable" in the sense that the history can't be reverted arbitrarily by someone other than yourself.

[5] https://www.coindesk.com/ethereum-executes-blockchain-hard-f...

> protocols monetized with market forces,

What do you mean by this? Older protocols like TCP weren't made operational for free, they were funded by many institutions, both public and private. Are you sure this is something new?

> fee-less payments that cannot be prevented

Cryptocurrencies aren't fee-less. There are transaction fees, exchange fees and volatility/liquidity risk. If you're in the US and want to send money (real money) to your friend in Japan, you must buy crypto (exchange fee), send it (transaction fee), and sell it (exchange fee and volatility or liquidity risk).

Finally, as you've seen with theDAO fork, crypto payments can be reversed, which prevents the transaction from being final. I'd love to see any research on how crypto payments can't be censored, though, if you have any.


This comment is just wishful thinking and stems from a false fact. There are HNers who were using "blockchain" even before Vitalik knew about bitcoin.

There is no innovation happening. Distributed is not decentralized. Ethereum's blockchain is a mathematical proof of it being mutable (and thus not a blockchain?). Tell me any "dapp" and I will tell you the point of centralization. (Before you even mention it's probably 'the origin of money' )


Well, it's already 10 years down the line with blockchain technology. Still waiting.

And, despite your silly claims of jealousy, the real reason blockchain technology now receives such scorn is precisely because for years after its inception it received absurd hype. What you're seeing now is the backlash against the utopian, almost religious praise of the technology when it first came out.

It's precisely because it's been 10 years and the reality hasn't nearly matched the early hype that people have cooled on the technology, and now want solid results instead of more empty talk. It's long past the point where telling everyone big things are coming is enough, it's time to show those big things.


> Well, it's already 10 years down the line with blockchain technology. Still waiting.

To be fair -- it's 10 years since Bitcoin. People didn't start innovating (widely) on blockchains until Ethereum came along and paved the way for generalized blockchains.

> It's precisely because it's been 10 years and the reality hasn't nearly matched the early hype that people have cooled on the technology, and now want solid results instead of more empty talk. It's long past the point where telling everyone big things are coming is enough, it's time to show those big things.

Excessive hype is normal and expected for any disruptive technology. AI took decades to become fruitful, the Internet took a few decades to become mainstream. We're seeing hype with cloud computing, VR, IoT, synthetic foods, and all of these have very vocal nay-sayers.

IMO, blockchains are revolutionary but also very primitive -- there's a ton of innovation in this space, and its spawned a variety of new areas for very interesting research (in trustless consensus, decentralized economies, stable currencies, algorithmic governance, and cryptographic applications.)

Of course, all of this is my opinion, and only time will tell.

(/me spent a decade working on distributed consensus infrasatructure at Google.)


This is that "utopian, almost religious praise of the technology" that's being talked about. What is actually disruptive about blockchain?

It's funny, all these technologies you listed are all ones that are hyped for hyped sake before having an actual use case. What actual, useful purpose does Blockchain provide? Like really. Apart from "a computationally burdensome way to hate the government."[1] Because all I ever see things like "blockchain-backed sex contracts" or "mangocoin - how IBM helped Walmart and farmers". I thought the most promising use case was when the Australian securities regulator announced they were going to put all the shares in the blockchain, but then I just realised that there's no actual concrete benefit from it - nothing they're doing can't be achieved by a simple, centralised database.

[1]: https://twitter.com/sarahjeong/status/951163739562061824?lan...


I think this might be because you live in a relatively trustful world of the west. Trust and privacy online will only become harder and harder to come by with how our current systems are set up.


> Excessive hype is normal and expected for any disruptive technology.

Plenty of hype for things that didn't make it though. Excessive hype isn't an indicator of success....

IMO, blockchains are worthless energy sucking devices whose only purpose is to take money from rubes and transfer it into the pockets of scammers. It's the worlds greatest platform for financial fraud and that is about it.


There were plenty of naysayers back then too.

Engineers love to play devil's advocate. Now that the posts/articles about cryptocurrency are becoming more frequent, those dissenting opinions are also becoming more frequent.


I have two problems with all existing crypto currencies.

1. Are they fundamentally safe or are the scams? 2. Will they get shut down by governments?

Both of those questions leave me feeling very open to risk. People willing to assume that risk before either issue has been resolved may reap whatever reward early adoption gives them. But if you build a business on the wrong tech you could get bankrupt or worse.

Most of the big dotcom 1.0 companies disappeared (or are dying before our eyes, RIP Yahoo). Maybe it is better to be a crypto 2.0 business.


Don't mix up crypto currencies with blockchain

Think pets.com vs the internet


I'm thinking more like SOAP vs. REST, except with the penalty possibly being outstanding debts in a currency that has been made illegal.


It’s more SOAP vs TCP/IP, the latter being Blockchain.


Cryptocurrency is the economic layer of a decentralized blockchain.


My "hate" has nothing to do with what you described but more on how the project is run.

It's no longer what it claims to be, and increasingly moving away from decentralization while at the same time the high priests creating propaganda that it's OK to be a "little bit centralized" for the "greater vision". Looks like most uneducated people genuinely believe this because they have no idea how Ethereum works underneath, and even many entrepreneurs who DO understand how Ethereum works are so invested in the ecosystem as to be blinded by their own interest, so "choose" to believe in it, but if they all stepped back for a minute and look at what has happened they will all realize this is not the platform they were excited about in the first place.

I generally hope the idea itself will work out, but doubt the current team and the entourage (also known as "ecosystem") around them will get there.


Please HN...

Can we approach this with an open-mind? These technologies have major promise. I do not see the use in being reactively closed mind and negative towards the efforts here. As an Ethereum - not a "crypto" - supporter, I feel brow-beat each time one of these threads comes up.

Let's take a deep breath, explore it, and try to help it realize its potential instead of instilling doubt and criticizing. This is a community for forwarding thinking builders...


I think it's well past that point. Now that there's serious wealth tied up in these projects and people have become millionaires and even billionaires on an unproven technology the general tech community is not going to give it the benefit of the doubt. This isn't just some experiment between crypto-nerds anymore. It's big business. It must stand up to the scrutiny now, and those that support it and have invested heavily in it should want that too.

I'm a bit ambivalent, there's clearly a lot of scams and way too high valuation. But something useful will likely come out of it all. And that something will likely be what can stand up against the backlash.


On a positive note sharding technology does have great promise because of the problems that it can potentially solve. With Bitcoin you can only validate a transaction by having all of the unspent transactions in memory [1]. With a sharded system you only need to have a subset of those transactions in memory.

At a fundamental technological level you are increasing the size of available memory by using the memory of other nodes. And there are ways you can query other nodes about their subset of unspent transactions by using a merkle tree. If you trust the root of a balanced merkle tree you only need O(log(n)) computations [2] to prove that a leaf belongs to it.

This sharding mechanism could be added to bitcoin where the root of the merkle tree of the unspent transactions is added to each block. Nodes on the network can store a sub-tree of arbitary size and validate part of each block. Whether this scheme would be robust to dishonest miners is still an open question though.

[1] http://gavinandresen.ninja/utxo-uhoh [2] https://gist.github.com/maaku/41b0054de0731321d23e9da90ba4ee...


> With Bitcoin you can only validate a transaction by having all of the unspent transactions in memory.

I'm somewhat sceptical of that claim, for the reason pointed out in the blog post, and I think you can get away with much simpler solutions for quite some time.

- You can use many heuristics to decide which UTXO data to keep in memory. LRU + deprioritizing addresses where the balance is too low to pay the current transaction fees seems like a good start. Research shows that there is a high percentage of stale data[0]. - With a block time as high as 10min, you can definitely spend 1ms on read latency for retrieving UTXO data from disk. - Even if you don't manage to retrieve UTXO data from disk quickly, that just means that the transaction missed its chance to be included in the current block. It could still be ready in time for the next one.

Overall, UTXO set growth is, of course, a problem since it increases centralization, but I don't think the problem of _in-memory_ UTXO sets is as big as it is often made out to be.

[0]: https://eprint.iacr.org/2017/1095.pdf


I feel the same way, I'm skeptical to post anything crypto related here anymore, meanwhile I'm busy "buidling" and keeping up with crypto / ethereum communities on discord


They've "showed promise" for almost a decade.

For almost a decade, evangelists have flooded tech forums promising to change the world, and for almost a decade they have failed to deliver.


Ethereum has been around for less than 4 years


I'm really really really not qualified but:

TLDR, split up a blockchain into a bunch of smaller blockchains sharded by something (let's say address), and then a different tier of nodes which only operate at the shard level can exist -- they call these "collations". collation-internal transactions can be managed by only the nodes in the collation, and they even lay out an initial proposal for how to do cross-collation transactions, leaning on merkle proofs for verification. They also propose making it transparent as a part of an auto-loadbalancing part of the blockchain so developers wouldn't feel any difference.

This paper was actually pretty interesting. It's pretty thoroughly researched, and the solution that they come up with (sharding) is actually pretty promising as a solution to the centralization + throughput issues of the blockchains up until now. I certainly learned a lot about blockchain systems by reading it, and more importantly learned the current problems with the state of the art and how this group plans to address them.

At this point, I also look at blockchain as a generic distributed database problem -- so I get the feeling these finndings will benefit the field some day. Surely something will come out of the research energy that's being dedicated to blockchain right now that will be beneficial to future researchers.


That's pretty accurate. One minor thing:

> developers wouldn't feel any difference

The details haven't been decided yet, but developers will need to deal with some kind of locking or "yanking" [1] mechanism in order to communicate with accounts on foreign shards.

[1] https://ethresear.ch/t/cross-shard-contract-yanking/1450


The section I particularly have a problem with is "How is the randomness for random sampling generated?" Everyone seems to think this is a solved problem but the reality is it is extremely difficult to solve. (The same problem exists for proof of stake schemes)

There is lots of complex cryptography, game theory and economics involved in working out if a robust solution really exists. The most commonly stated solution is something called a "Threshold Signature Scheme". Threshold signatures schemes (and threshold cryptography in general) have two main problems.

Firstly they don't scale well. As you add participants you need lots of extra rounds of communication and computation.

Secondly they aren't resistant to Sybil attacks. The proposed way around this is to link participation to a stake of coins (yep back to proof of stake) such that launching a Sybil attack requires lots of coins. This is not trivial however because of the "Nothing at Stake" problem where stake can be used to participate in many forks of a scheme at the same time.


Ethereum POS is already running on a test net, you're welcome to go over there and see what happens when you try to use your stake on "many forks at the same time". Spoiler: you won't be able to do so without losing your stake very quickly.


Well not quite. The test net is actually a hybrid of proof of work and proof of stake [1]. Proof of work is used as the main decider in the case of a fork, however if long forks are seen then checkpoints validated by proof of stake are used.

The slashing function is in the contract [2] but there is actually no incentive to duplicate votes on the testnet since there is only rewards for voting (block rewards are given to proof of work miners).

[1] https://github.com/ethereum/casper/blob/master/IMPLEMENTATIO...

[2] https://github.com/ethereum/casper/blob/master/casper/contra...


Don't totally follow your argument, but if you're saying that we won't have a full understanding of the POS economic guarantees until we have empirical evidence from a "live network" with significant rewards at stake I agree 100%.


If PoS fails, it will fail as a black swan event, after the system became "too big to fail", not at the level of mere testnet.


Another approach is to have several parties (perhaps every block creator) contribute a number. Then we concatenate the numbers and feed them through a delay function, which should take everybody a certain minimum amount of time to compute, regardless of computational power. The result of this function becomes the random seed for some future block.

An attacker cannot predict how their number will affect the result, since they would need to compute the delay function first. By the time they finish the computation, they'll have missed the window of opportunity to contribute a number. (They could add their number on a very old fork, but other clients would not accept forks beyond a certain age. This is important anyway to prevent stake-bleeding attacks.)

A obvious candidate for a delay function is an iterated hash function like sha3^10000. A few researchers are looking into "verifiable" delay functions [1], so that not every client has to do the computation; most clients can download the result along with an easily-verified proof (perhaps using a SNARK) that the result is correct. Besides saving some computational resources, this would help new nodes catch up to the network quickly, without needing to compute all the historical seeds.

[1] https://www.youtube.com/watch?v=qUoagL7OZ1k


Easy way out: Use the NIST randomness beacon. Now, if only half the pull of cryptocurrencies wasn't the rampant libertarian dream of it...


Interestingly Ethereum has hit in a sweet spot: tech crowd that learns EVM quickly and enjoys playing with it (like a new raspbery pi) & has sizable spare cash to invest in ether/erc20. Bitcoin is nowhere as fun to learn, the VM is quite boring. I wonder if any other blockchain can hit into this sweet spot.


I personally think that sharding is going to work great because for starters 100% trustlessness was, in my opinion, never a real requirement. If it was then nothing in our society would work.

People are going to look back and say that the HBO show 'Silicon Valley' was 'prescient', but really they are just catching up to developments in Ethereum and similar projects.

Long live the real PiperNet (well they have some significant differences but also some similarities).


Good for them for nerding out with all the esoteric tech to solve the decentralization problem.

But evidently all this means nothing if it's possible to mutate the blockchain itself through social and political engineering.


But if that mutation needs majority (from miners or stake holders) how is it any worse than a democratic country deciding to change a law or print money or whatever?


Wasn't the main part of the allure to blockchain/crypto to be better than democratic societies and institutions? After all that's what they are trying to replace right? Governments can't be trusted and all that?


Presumably it's a lot harder for a group of evil-doers to get a country into changing laws or printing more money. Your opinions may vary. I have my faith in (some) countries.


It's far worse because it lacks any kind of controls or process to make sure everybody gets a fair shake. A good government answers to all citizens. Blockchains answer to those with the most hash power (i.e. $$$$).

You are effectively saying you'd prefer mob rule over due process.


'FAQ' seems like the wrong term. I doubt that "What are some trivial but flawed ways of solving the problem?" is frequently asked...


Eh, having had quite some discussions on the topic, a lot of the "flawed ways" come up as "why don't we just do this?" suggestions.


Yeah, this is more like the Socratic method but one-sided and with leading questions. Being a method for uncovering the truth, I guess that's OK if the question you're trying to investigate is "What is the proposal?".


I think maybe you could improve on this sort of approach by weakening the invariants. Instead of trying to make sure that only valid "collations" get on the main chain, let anyone willing to pay for it put collations on the main chain but define the first honest one on the chain to be valid. You can't tell which ones are valid at a glance but you can with O(shard history) resources.


If you can figure out the details on that, they'd probably be interested.


Having just attempted to read and comprehend this, my total conclusions are: (a) I have a headache (b) I have absolutely no idea if this is going to work.

(if anybody has more useful conclusions, I'd love to hear them, most of the threads seem to be alternating "blockchain yay" and "blockchain boo" comments, neither of which tend towards being desperately informative)


I feel the same, the arguments about blockchain is one of the more fruitless one about technology(scams and over-hype makes it worse). Meanwhile the doers is trying to figure out how to make it useful. Once some infrastructure and successful examples are done(For example block chain collection done right or oracles that publish user or sensor data for other smart contract to use at a cost). We will have much more stuff to start arguments from.


I wonder what effect PoS will have on the Ethereum ICO system. If coins are needed to stake and mine, large holders might have more incentives to hold Ethereum instead of any assets?


Sharding will enable Ethereum to scale without compromising on the decentralised and permissionless nature of it's blockchain implementation. Really excited about this!


What's the need for a high transaction rate? "10,000 TPS". Right. Nobody uses cryptocurrencies for small transactions any more. It's just a speculative investment.

Bitcoin being stuck at 7 TPS is embarrassing, but if they could get up to a few hundred TPS, that would probably be adequate.


The end game goal of Eth is to become the main way to process transactions globally, overtaking Visa et al.

Not only are cryptos still used for transactions (the dark web marketplaces are still used after all) but they have ambitious goals for widespread adoption through dapps and larger financial integrations.


Ethereum is also a DAPP platform.

In theory you could build a decentralized Twitter/Facebook using Ethereum but without 10k+ TPS there is no practical way to get enough users using it.


Honest questions why would you need a blockchain for this? I can understand for money transactions/contractual signing but for Facebook? Does anyone really care that there is a ledger for Facebook?


Sharding: Putting the trust in trustless

/s


TBH part of what makes me so bullish on Ethereum is how much of an emotional reaction it produces, both for and against the technology, all the while it continues to gain mindshare and utility.


I've had a couple friends try to enroll me into their ponzi schemes. They were really emotional both during the sale and after I remained in "no". I think emotions are just the communication of the base animal part of people. No need to imbue it with prescience or exceptional intelligence.


Ponzi schemes are terrible. Many ICOs aren't much more. However, blockchain enables digital scarcity. That's an interesting piece of technology right there. Using it as a state-machine is a very interesting application of it. IMHO many negative emotional responses come from scamming applications, many positive ones from new possibilities created by the technology, some of both are just due to the transformative nature of its application.


> However, blockchain enables digital scarcity.

Eh? Scarcity? Ever looked at all the forks of bitcoin?

Blockchain absolutely does not enable scarcity. This has been proven time and time again...


You say that as if Bitcoin forks are unquestionable proof of increasing supply, which many people will disagrewle with. I consider them to be completely separate products, seeing as the forks are not backed by Bitcoin's hashpower.


It doesn't matter what someone "sees" them as, the fact remains that forks are possible because the scarcity is an artificial illusion.


Sorry for being ridiculously inappropriate, but the cryptocurrency topic calls for me to leave this more appropriate title here:

Ethereum Sharting FAQ.

Have a nice day!




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: