I don't want to be anti-blockchain, but I have yet to come across any problems that are solved better with a blockchain than with any other traditional technology - except for the original one, a distributed electronic currency.
I don’t understand why you would be anti- something so new at all. To me it’s preposterous that you would evaluate a new technology based on how many problems it and only it can solve. Don’t you get that new technology has an incubation period where its fundamental challenges must be overcome?
What problem could early steam powered cars solve that horses couldn’t solve better? I doubt the earliest ones performed better than the animal based solutions for ANY specific use.
Yet writing off automobiles at that stage would’ve been a mistake only obvious in retrospect.
Especially for something as truly bizarre as Ethereum, why write it off at this early stage?
How long do we have to wait before we can consider blockchains not so new? Bitcoin is 10yo, the whole field has received literal billions in founding in the past few years (anybody with a good looking website and a three page whitepaper could make millions in an ICO last year). Furthermore anybody coming up with an actual innovation in the field is sure to make millions so the incentive to innovate is huge.
So, where are the results?
Your car analogy doesn't add up, for quite a while horses were undoubtedly more convenient than automobiles. Then automobiles improved and took over. We have a similar situation with electric cars today.
Blockchain technologies are not a few cycles of optimizations away from being able to compete with PostgreSQL, they work fundamentally differently and they have fundamentally, per design, strong limitations about how efficient they can be. That's the cost of trustless decentralization.
To overcome these limitations you don't need "a blockchain but more better", you need a fundamental breakthrough in the way these things work. I suppose you could hope for such a breakthrough but that's akin to hoping that we're going to figure out cold fusion and solve the energy crisis. Sure, it can happen but it's not unreasonable to be skeptical of it.
You're praising innovation blindly. Not all innovation is good, and many supposed innovators are crocks. I understand blockchains enough to where I could build one of my own (note that I say blockchain, and not cryptocurrency) and I really don't see much use for this particular datastructure outside certain niche applications.
What are these truly innovative uses for a nested chain of data blobs whose contents include a hash of the previous link?
He isn't praising anything. He is simply saying not being anti-something that early. The world isn't all black and white, there's nuance, not being anti-something doesn't means praising it.
I don’t understand why you would be anti- something so new at all.
>[...]
> why write it off at this early stage?
What if there's actually nothing good that we can do out of it but you aren't there to talk against it? Well it will die naturally. What if there's actually something good that we can do with it but there's a bunch of anti-something people? Well that add unnecessary friction to something good being made and that just make you a jerk adding it.
A (potentially distributed) datastructure of some kind, which resembles a linked-list, and ensures its integrity using some sort of cryptographic hash of the previous link such that changing any link but the most recent invalidates the whole chain. Literally -- a chain of (data) blocks.
A non-engineering description of a "blockchain" should be considered to be misleading.
edit - also, I'd take cryptocurrencies more seriously if they could stay out of the headlines for a while
Is utterly funny to read people's opinions bashing against blockchain technology and when asked about a simple definition they fail to mention the use of consensus protocols which was, already 10 years ago, the huge innovation that lead to this revolution.
Especially for something as truly bizarre as Ethereum, why write it off at this early stage?
If a new invention is truly transformative, it will thrive even after long doubt. At this stage the public, forced demonstration of cryptocurrencies deserve to be criticized no matter it will be successful or not. (I don't want to criticize general cryptocurrency researchers of course, because, well, they are just studying.)
There are so many advantages to early cars, are you kidding me? They don’t eat hay and shit for one.
Is there anything that your argument wouldn’t work for? Do you truly have to wait a decade before making any judgements about the hype around a technology? When will we be able to say if smart contracts have been overhyped?
They eat coal/gasoline which is probably more expensive. Also they worked on almost no terrain at all compared to horses and constantly broke down.
The only reason they would even make sense to pursue would be because of how successful the train was. Similar to how this bitcoin stuff makes more sense when you look at the success of the internet/paypal/amazon.
You know you can research this stuff right? Gasoline was a waste product in the refining of crude oil (specifically, into Kerosene [itself a replacement for whale oil] they were making anyways). It used to be considered garbage. [1] It really only became valuable once they realized they could power cars with it. They used to dump it in rivers, making them catch fire. [2]
In what - concrete - way is bitcoin related to the success of Amazon or the internet? Please be specific and don't hand-wave if you're going to make such bold claims.
I am not anti-blockchain, there are applications where it makes sense. But 90% of the time when I hear about a blockchain related application its a regular application which just so happens to use the blockchain as storage because its a buzzword.
It feels like the world suddenly discovered the hammer, and now we think everything is a nail
Like what? Everyone seems to hand-wave this, but nobody in the full decade long history of bitcoin and blockchain has come up with a single good use case that isn't better solved by MySQL.
Illegal gambling and illegal transactions tend to be good Bitcoin usecases that are trickier to solve with MySQL due to trust issues. Now, you may not be particularly enamored with illegal activity, but it is the one use case where Bitcoin really excels.
MySQL doesn't guarantee message integrity, so anything involving shared data among separate entities that could be attractive for manipulation would be better solved with blockchain tech than MySQL.
The web wasn't invented until the internet was more than a decade old.
This narrative that "Internet time" transforms the world every other year should have died in the 90s. Real work takes time.
Building the infrastructure for the Internet took decades. Even after it won, most people used modems over badly insulated copper wire for the better part of ten years.
Bitcoin is a “product” like the web and it’s existed for a decade. Blockchain is just hashing, infrastructure that’s already existed for decades. Nobody has found a good use for either yet. They may. But that doesn’t mean any particular cryptocurrency should have any value. Or that we should particularly value the blockchain yet.
The reason nobody has found a good use for block chain yet is that the only time it offers any benefits over a traditional relational data store is when you cannot trust anybody. The truth is you can trust someone almost all the time. And even if you can’t really, it might be more efficient to just pretend. The truth is that trust is a substantial optimization and it eliminates vast quantities of waste and inefficiency in the proof of work model. That’s why people don’t want blockchains.
A really clean way of expressing this idea is that when you deposit a paper check at the bank they don’t check the signature. They don’t check the signature because they trust you, or rather it’s more efficient to just assume the money is being deposited legitimately. The amount of effort signature verification takes makes it more expensive than just eating the fraud losses. The same is true in many of these block chain powered proof of work situations. The cost of computing the proof of work outweighs the cost of managing fraud. Not to mention all the ridiculous fraud perpetrated on this unregulated market but that’s neither here nor there.
But no one person or company made billions off the invention of the web.
I wonder if investors are pushing developers away from infrastructure plays, towards consumer apps and sub-platforms.
I imagine that's happening in VR. I hear a lot about headsets, games, medical applications, etc, but I haven't heard anyone get excited about research into better Fresnel lens production.
USD is already a mostly electronic currency. I’ve done some reading, and I have yet to come across a compelling advantage for cryptocurrency cum currency for the purposes of legitimate transactions.
It’s mostly useful for people whose social networks aren’t fully enfranchised in the U.S. financial system. For people within that system your analysis is correct.
Interesting. Can you expand on what you mean by "people whose social networks aren't full enfranchised in the U.S. financial system?" The phrase has me thinking of this: https://en.wikipedia.org/wiki/M-Pesa
You could just use any free kind of canary instead, access to traps is easy to notice without any need for a complex external dependency or wasting big amounts of money.
Comments
I don't want to be anti-blockchain, but I have yet to come across any problems that are solved better with a blockchain than with any other traditional technology - except for the original one, a distributed electronic currency.
I don’t understand why you would be anti- something so new at all. To me it’s preposterous that you would evaluate a new technology based on how many problems it and only it can solve. Don’t you get that new technology has an incubation period where its fundamental challenges must be overcome?
What problem could early steam powered cars solve that horses couldn’t solve better? I doubt the earliest ones performed better than the animal based solutions for ANY specific use.
Yet writing off automobiles at that stage would’ve been a mistake only obvious in retrospect.
Especially for something as truly bizarre as Ethereum, why write it off at this early stage?
How long do we have to wait before we can consider blockchains not so new? Bitcoin is 10yo, the whole field has received literal billions in founding in the past few years (anybody with a good looking website and a three page whitepaper could make millions in an ICO last year). Furthermore anybody coming up with an actual innovation in the field is sure to make millions so the incentive to innovate is huge.
So, where are the results?
Your car analogy doesn't add up, for quite a while horses were undoubtedly more convenient than automobiles. Then automobiles improved and took over. We have a similar situation with electric cars today.
Blockchain technologies are not a few cycles of optimizations away from being able to compete with PostgreSQL, they work fundamentally differently and they have fundamentally, per design, strong limitations about how efficient they can be. That's the cost of trustless decentralization.
To overcome these limitations you don't need "a blockchain but more better", you need a fundamental breakthrough in the way these things work. I suppose you could hope for such a breakthrough but that's akin to hoping that we're going to figure out cold fusion and solve the energy crisis. Sure, it can happen but it's not unreasonable to be skeptical of it.
You're praising innovation blindly. Not all innovation is good, and many supposed innovators are crocks. I understand blockchains enough to where I could build one of my own (note that I say blockchain, and not cryptocurrency) and I really don't see much use for this particular datastructure outside certain niche applications.
What are these truly innovative uses for a nested chain of data blobs whose contents include a hash of the previous link?
He isn't praising anything. He is simply saying not being anti-something that early. The world isn't all black and white, there's nuance, not being anti-something doesn't means praising it.
What if there's actually nothing good that we can do out of it but you aren't there to talk against it? Well it will die naturally. What if there's actually something good that we can do with it but there's a bunch of anti-something people? Well that add unnecessary friction to something good being made and that just make you a jerk adding it.
You see how it's better?
What defines a blockchain for you?
I agree that a token based currency is not required, I'm just interested in what people define a blockchain as for intellectual curiosity.
A (potentially distributed) datastructure of some kind, which resembles a linked-list, and ensures its integrity using some sort of cryptographic hash of the previous link such that changing any link but the most recent invalidates the whole chain. Literally -- a chain of (data) blocks.
A non-engineering description of a "blockchain" should be considered to be misleading.
edit - also, I'd take cryptocurrencies more seriously if they could stay out of the headlines for a while
Is utterly funny to read people's opinions bashing against blockchain technology and when asked about a simple definition they fail to mention the use of consensus protocols which was, already 10 years ago, the huge innovation that lead to this revolution.
Consensus protocols are nothing new. The novel part of blockchains is the actual blockchain
https://en.wikipedia.org/wiki/Paxos_(computer_science)
You're thinking of a distributed ledger, which combine blockchains with consensus protocols
I prefer the "proof of work != blockchain" comments when people include the most common consensus protocol in their simple description of blockchain.
If a new invention is truly transformative, it will thrive even after long doubt. At this stage the public, forced demonstration of cryptocurrencies deserve to be criticized no matter it will be successful or not. (I don't want to criticize general cryptocurrency researchers of course, because, well, they are just studying.)
There are so many advantages to early cars, are you kidding me? They don’t eat hay and shit for one.
Is there anything that your argument wouldn’t work for? Do you truly have to wait a decade before making any judgements about the hype around a technology? When will we be able to say if smart contracts have been overhyped?
They eat coal/gasoline which is probably more expensive. Also they worked on almost no terrain at all compared to horses and constantly broke down.
The only reason they would even make sense to pursue would be because of how successful the train was. Similar to how this bitcoin stuff makes more sense when you look at the success of the internet/paypal/amazon.
You know you can research this stuff right? Gasoline was a waste product in the refining of crude oil (specifically, into Kerosene [itself a replacement for whale oil] they were making anyways). It used to be considered garbage. [1] It really only became valuable once they realized they could power cars with it. They used to dump it in rivers, making them catch fire. [2]
In what - concrete - way is bitcoin related to the success of Amazon or the internet? Please be specific and don't hand-wave if you're going to make such bold claims.
[1] https://knowledgenuts.com/2014/02/24/gasoline-used-to-be-con...
[2] http://www.attendly.com/rockefellers-unconventional-approach...
The internet and amazon displaced the old business models with e-commerce. Bitcoin could do that with money/gold.
This is similar to train->car.
I am not anti-blockchain, there are applications where it makes sense. But 90% of the time when I hear about a blockchain related application its a regular application which just so happens to use the blockchain as storage because its a buzzword.
It feels like the world suddenly discovered the hammer, and now we think everything is a nail
Like what? Everyone seems to hand-wave this, but nobody in the full decade long history of bitcoin and blockchain has come up with a single good use case that isn't better solved by MySQL.
Illegal gambling and illegal transactions tend to be good Bitcoin usecases that are trickier to solve with MySQL due to trust issues. Now, you may not be particularly enamored with illegal activity, but it is the one use case where Bitcoin really excels.
That’s a good use case but it doesn’t feel like a good flagship product to spur mass adoption haha
MySQL doesn't guarantee message integrity, so anything involving shared data among separate entities that could be attractive for manipulation would be better solved with blockchain tech than MySQL.
It's not new, BTC is a full decade old. At this point the internet had a billion users.
The web wasn't invented until the internet was more than a decade old.
This narrative that "Internet time" transforms the world every other year should have died in the 90s. Real work takes time.
Building the infrastructure for the Internet took decades. Even after it won, most people used modems over badly insulated copper wire for the better part of ten years.
Bitcoin is a “product” like the web and it’s existed for a decade. Blockchain is just hashing, infrastructure that’s already existed for decades. Nobody has found a good use for either yet. They may. But that doesn’t mean any particular cryptocurrency should have any value. Or that we should particularly value the blockchain yet.
The reason nobody has found a good use for block chain yet is that the only time it offers any benefits over a traditional relational data store is when you cannot trust anybody. The truth is you can trust someone almost all the time. And even if you can’t really, it might be more efficient to just pretend. The truth is that trust is a substantial optimization and it eliminates vast quantities of waste and inefficiency in the proof of work model. That’s why people don’t want blockchains.
A really clean way of expressing this idea is that when you deposit a paper check at the bank they don’t check the signature. They don’t check the signature because they trust you, or rather it’s more efficient to just assume the money is being deposited legitimately. The amount of effort signature verification takes makes it more expensive than just eating the fraud losses. The same is true in many of these block chain powered proof of work situations. The cost of computing the proof of work outweighs the cost of managing fraud. Not to mention all the ridiculous fraud perpetrated on this unregulated market but that’s neither here nor there.
But no one person or company made billions off the invention of the web.
I wonder if investors are pushing developers away from infrastructure plays, towards consumer apps and sub-platforms.
I imagine that's happening in VR. I hear a lot about headsets, games, medical applications, etc, but I haven't heard anyone get excited about research into better Fresnel lens production.
Well, you need a use case.
USD is already a mostly electronic currency. I’ve done some reading, and I have yet to come across a compelling advantage for cryptocurrency cum currency for the purposes of legitimate transactions.
It’s mostly useful for people whose social networks aren’t fully enfranchised in the U.S. financial system. For people within that system your analysis is correct.
Interesting. Can you expand on what you mean by "people whose social networks aren't full enfranchised in the U.S. financial system?" The phrase has me thinking of this: https://en.wikipedia.org/wiki/M-Pesa
They probably mean people who carry out illegal transactions.
From another thread:
“Code bounties are a great use-case for cryptocurrency.”
https://news.ycombinator.com/item?id=18226000
How is it any better than conventional tools such as Patreon or Flattr?
Why?
Without reading, a use case could be to sprinkle some private keys in your “secure” code and “secure” database.
If it gets leaked/hacked, there will be temptation to cash it out. Follow it and get quick notification that your security has been breached.
You could just use any free kind of canary instead, access to traps is easy to notice without any need for a complex external dependency or wasting big amounts of money.