There's No Good Reason to Trust Blockchain Technology (2019) | Hacker News Reader

There's No Good Reason to Trust Blockchain Technology (2019)

wired.com

120 pointslouwrentius172 comments

Bruce's take can be summarized like this: with a blockchain, you're choosing to trust code and cryptography instead of a person. And when immutable code does the wrong thing, you can't trust them to fix it like you could a person.

This is a good take, and underappreciated - but there's a larger point being missed.

In high-social-trust situations, there are fewer benefits to blockchains. In low-social-trust situations, people aren't a good option. Bockchains provide an entirely new and welcome option for that scenario.

Relatedly, there are different kinds of distrust: mistakes and attacks are one, but platform risk is another that blockchains can alleviate. Putting it another way: I trust Twitter/YouTube/PayPal to generally work without issues, but I don't trust them not to change their offering in an exploitative way or become rent-seekers.

> In high-social-trust situations, there are fewer benefits to blockchains. In low-social-trust situations, people aren't a good option. Bockchains provide an entirely new and welcome option for that scenario.

How does this actually work in practice? I've never seen an actual non-convoluted / non-far-fetched case where this would apply.

For one: irreversible payment. Dealing with lots of scammers as a seller? They can PayPal dispute. They can chargeback. But if they send you cryptocurrency, the money is guaranteed.

This doesn't prevent scams in the other direction, but it's something. This is a common problem many independent sellers face when doing legal and legit business.

> This doesn't prevent scams in the other direction

Which is probably the majority of the actual scam cases. Sellers scamming buyers.

So I don't think this is a good case at all.

I could not disagree more, as someone who operates a niche ecommerce business. The longer I exist, the more reputation I have. But every single customer I sell to is a new risk, and the credit card companies fob all the risk onto me. Crypto payments (which I accept as an option) are great because I have no risk of a charge back a month after I shipped. The risk of being scammed is so much more about anonymous customers than it is of businesses, who need repeat business and word of mouth.

The problem is that the risk is borne by honest business owners and honest customers. Scammers exist in both directions and the scammers typically bear none of the risk.

Regarding businesses specially, someone operating one or more scam online retailers is actively trying to suppress word of mouth. They also don't care about repeat business because their profit margin is so high. Their process may even specifically include periodically cutting ties with their previous name, websites, vendor accounts etc when victim ill becomes unmanageable.

If you choose to purchase from sketchy merchants, you have all sorts of risks regardless of what payment infrastructure is used.

As a business that wants to last a long time and attract customers, reputation is everything. People buy from large, established companies for this reason. Crypto payments help all sellers by eliminating an enormous problem of scam buyers using stolen credit cards.

Well, it is a good case for an honest merchant, that is dealing with people trying to scam them.

I don't know if that's true at all, and I suspect it's false. While both types are not uncommon, I think sellers being scammed is the more common scenario.

Even if it wasn't, though, how can you say this isn't useful? Even if 99% of scams were from sellers, the fact that this trivially and provably solves scamming from buyers is amazing. Saying this isn't a good case because cryptocurrency doesn't miraculously universally eliminate the possibility of any fraud from any party in any scenario is really setting a very high bar.

And with things like Ethereum smart contracts, you theoretically can prevent fraud from all parties in some cases, and defer all risk to the implementation of the contract. Or do you consider this a bad case as well, because a contract could possibly be flawed and "the majority of fraud isn't actually due to human deceit"?

I remember when eBay operated on people sending checks to each other. Good times.

So it solves this by superseding consumer protections? What does blockchain add here over direct payments, or cash, or even some payment broker with more seller friendly terms? I can't help but assume that regulation will reinforce these protections.

Essentially, crypto's major benefit is skirting regulation because its new and that can't last forever.

How do you propose a direct payment or cash without giving a buyer your bank account number or address, which you may not necessarily want to do for privacy and security reasons? And mailing cash isn't very efficient or practical.

Payment brokers are always going to be a point of failure. But also, it's about pragmatism: can you name an existing payment broker that fits the criteria? If not, what option are they left with other than to use cryptography as their broker?

But, yes, it's often about avoiding certain kinds of regulation. Regulation isn't inherently good or bad. Some regulation is good in some contexts and some is bad in some contexts. Cryptocurrency lets you enforce your own absolute regulation for cases where legal regulation isn't able to easily help you (like in this scenario).

Blockchain partially solves the problem of laundering illicit funds. The most popular use case I’ve seen is receiving ransoms.

My bad. I've should have added non-criminal. I'm sorry.

A criminal use is still a use. Some percent of gun sales are for criminal purposes. This counts as market demand just the same as any other sale. Precisely measuring how valuable this is to criminal enterprise is an interesting question.

My main area of skepticism regarding this proposed use case is that a blockchain is an immutable public ledger, and thus with enough sleuthing, it is probably quite traceable. Indeed many criminals have later been apprehended through monitoring bitcoin transactions.

If it is a non criminal situation, you can generally use traditional methods to establish trust, such as using courts to sue people who rip you off, or using escrow agents, etc.

The world has had thousands of years to come up with solutions to trust problems during transactions. They are solutions that work. They work for criminals too, but its a lot harder to use them, especially over the internet, because everyone providing the services goes to jail if they are transacting with criminals.

> such as using courts to sue people who rip you off,

Unless the whole transaction is anonymous, you can do the same with bitcoin.

> or using escrow agents

Which costs you extra.

Not making a value judgement, just pointing things out.

> you can do the same with bitcoin.

Yes of course - but the question was what value does bitcoin add over normal money, not /name things you cannot do with bitcoin/.

In Belarus, it is criminal to donate to the opposition party or to the wounded protesters.

A fucked society is not fixed with bitcoin

Lol! That is suspiciously specific.

You didn't need to say illicit funds. It's useful to move money without having to use a bank or government money.

It may be useful, but the generic case of transfering money is generally (but not always) not an example of a low trust situation.

Paying off ransomware software is.

And we can see this in the fact that buying goods and services from traditional (not darknet) vendors is still very hard with bitcoin. Paying randsomware in bitcoin is standard.

For most goods and services there are desirable properties to intermediaries in terms of fraud detection, refunds, etc. But it is ultimately reliant on the ultimate source of trust deriving from governments and banks.

If you are a bank or government you have to look out for yourself, and so a fair amount of blockchain hype specifically addresses the low-trust transactions they encounter. There is some quiet adoption happening globally, and mostly invisibly to consumers.

The side effect, of course, is that it's easy for anyone to do bank-like things with the tech, including ransomware.

Voting and currency are 2 real world situations.

Bitcoin isn't convoluted or far fetched even if you really want to disagree.

It's nuclear chemistry proof (unlike gold), and outside the reach of government.

Its no more outside the reach of government than you are.

Voting sounds like a terrible use case for traditional bitcoin style blickchain. Am i missing something?

> In low-social-trust situations, people aren't a good option. Bockchains provide an entirely new and welcome option for that scenario.

We've been handling these set-ups very well for millennia. You find a trusted intermediary. These days, for better or worse, globalization has created more and more of them.

There is even a word for it: escrow

Indeed, the process is quite simple.

You want to do business with Steve, but you don't trust Steve. Cool. Do you and Steve mutually trust someone? Awesome, you've found an escrow. If you can't find someone who trusts both you and Steve, that's a great sign you shouldn't do business with Steve.

No amount of magic Blockchain pixie dust is going to change that there are bad actors out there.

It’s more for when you already trust Steve, but you don’t trust your national currency, the credit card companies, and other available “escrows”. Why force use of an escrow when you trust your counterparty ?

> Why force use of an escrow when you trust your counterparty ?

Why use a trustless payment method when you trust your counterparty and waste a month's worth of electricity?

Because my counterparty may not be able to spend my IOU elsewhere.

This situation feels pretty contrived, do you have a concrete example, especially one which you've seen personally or been personally involved with?

not personally. but imagine you are the king of island A, and want to engage in a transaction with king of island B. Or you want to split up your estate amongst your children.

I get that, but I firmly believe that solutions, certainly ones as highly touted here, should operated in the problem domain. We've yet to identify, in 11 years, a concrete problem that blockchains are better at solving than any other solution.

But what if your business is ransomware or selling narcotics, and it's hard to find a trusted intermediary?

For everything else, there's wire transfers MasterCard jingle plays.

Those intermediaries can be moderately to very expensive. Look at how much value credit-card interchange skims.

Blockchains offer a technological solution to a social problem- and threaten many "trusted brokers" with replacement with near-zero marginal cost software.

The fact that a single Bitcoin transaction requires 561kWh of power (enough to power an average US household for 19 days) -- and produces 94g of e-waste -- means that classical intermediaries are a lot less expensive in the majority of cases. [1]

Using the world average of $0.14/kWh means that a Bitcoin payment costs (or rather extracts in welfare) $79. This fee is socialized across all users in the form of block rewards, and as they diminish, end-users will have to bear the whole cost or the hash rate will drop off precipitously eliminating the security guarantees. This is on top of the $3.392 fee for processing the transaction on-chain.

At a break-even of $82.39 per BTC payment, a credit card transaction would have to be at least $27,500 in Europe (based on their 0.3% interchange cap) and $4,551 in the US (based on the average interchange rate of 1.81%). ACH payments cost banks $0.11 US.

Trusted brokers aren't going anywhere, and IMO, there's no such thing as a technological solution to a social problem.

[1] https://digiconomist.net/bitcoin-energy-consumption/

These things are true at present but are they true in the limit? Bitcoin is the first blockchain technology to see adoption and has a present market cap of over $200B despite scaling incredibly poorly.

Trusted brokers won't disappear but there's a big difference between being the only game in town and filling a niche in an ecosystem.

Technology makes social solutions irrelevant all the time. How many interactions intermediated by people no longer exist because they are now intermediated by software and platforms?

We've had 11 years of innovation in blockchain technology and so far, nothing has improved on the status quo. 11 years ago was right around when the first iPhone came out, for reference. In my opinion, that's because you can't solve trust with magic pixie dust. No matter what, trust represents a huge optimization. That means anything you try and do to reduce trust will introduce massive quantities of friction. Trust is good and should be embraced.

If something better comes along, great, but I've seen no evidence in the last 11 years that would indicate anything's at risk of coming down the pike. I've got no interest in championing a technology that hasn't budged a single iota (heh) in 11 years. If it does, I'll eat my... well. You get the joke.

[edit] note that market cap doesn't mean anything for a currency. There's no market cap of US dollars, because dollars have a fixed value relative to dollars. If I exchange my dollar for a dollar, the market cap of dollars is equal to the number of dollars in circulation. End of story. You value other things relative to currency to obtain their market cap.

The market cap of BTC is 21MM BTC. Less, given the huge quantities that go missing all the time.

Proof of stake vs proof of work?

> These things are true at present but are they true in the limit?

In the limit no more BTC can be obtained by mining, so transactions will be more expensive, because the computational cost will have to paid for by the transactor.

And the computation required will be much larger than now, requiring more energy, and, given the end of Moore’s law, will likely be vastly more expensive.

It is doomed to fail.

It may be doomed to fail, but it might be useful until it’s failed by something better. And that may be a long time.

Right, but it isn't. We've yet to find a single use case better served by crypto than traditional means, in 11 years, that is legal.

It's useful at burning incredible amounts of energy doing essentially nothing useful.

It currently uses about 0.25% of the world's electricity, which, for a market cap of ~100B, is an astounding waste.

Global currency trading, at $6+T per day, at that scale, would use 15% of global world electricity usage, which is more than any country besides US and China.

Probably more, because it's designed to be actively anti-efficient. The more you try and use it the less efficient it gets. It might well require more than Earth's total power generation capacity to do so.

Anti efficiency is a feature in other places as well: password complexity, safety procedures, memory safety, functional programming (immutability), checks and balances in government.

Electricity is renewable. Waste is unused supply. there is more demand for bitcoin than supply.

I usually agree with most of your analysis on economics, except on bitcoin. Hopefully you will come around on this topic as well.

>Electricity is renewable.

That makes no sense. Electricity is almost always turned into some other energy form upon use, and once there, it's forever lost energy. This is basic physics. Even transmission loses energy. Electricity is not "renewable". That statement makes no sense.

Electricity can be generated by renewable resources, but used electricity usage ultimately becomes heat. That entropy increase cannot be undone without a larger expenditure of energy to recapture. Basic physics. It's lost.

Next, the majority of electricity is generated by burning fossil fuels. "Renewable" also has externalities: solar cells produce very toxic waste on construction and decommission, nuclear produces spent fuel that is highly radioactive for thousands of years, hydro requires massive cement outlays which has serious environmental costs, and even wind farms create a decent amount of waste on construction and decommission.

>there is more demand for bitcoin than supply.

This means nothing. There is more demand for nearly everything on the planet than supply. That does not mean doing inefficient things is suddenly wise.

>I usually agree with most of your analysis on economics, except on bitcoin.

Thanks. The problem is I won't come around, since BTC is one of the dumbest solutions economically created in modern times. Man spent millennia making errors on how to deal with money, currency, and macro economies, and BTC has skipped those painful millennia and recreated about the worst features in every aspect. It's what you'd expect from a programmer detached from economics and history.

It's deflationary, it cannot account for shocks, it's about the most volatile and speculative place to put money over the past decade, it's not useful as a currency due to volatility, it's not useful as a store of value since it's unpredictable, it allows incredible theft and loss (check what % of BTC has been stolen or lost over time - it's astounding), it's still subject to all sorts of bad actor problems, and on and on.

It's hacker cool, but economically absolutely retarded. There is no question about that.

You may have drank the kool-aid, but economists have not, and for good reason - they know how money, economies, and exchange works, and are aware of all the errors that mankind has grown out of that BTC blissfully embraces.

You should trust a money system created by a hacker as much as you'd trust a cryptosystem created by a painter. They're both as well designed.

Edit: after I wrote this, I googled to check if there's a good page on the terrible economics of BTC. As usual, Wikipedia did not disappoint. Check the section labeled "Views of Economists" [1] to see how many Nobel prize winning economists have called it basically crap. Not one in that list had a good thing to say on BTC.

This is the point where BTC zealots apply Dunning-Krueger and view themselves as economically wiser than this many real experts.

Edit 2: There is a neat poll of a decent chunk of the world's top economists, and I never looked there to see what they have been asked about BTC. There are hundreds (thousands?) of cool questions put to this panel, and quite often when I want a feel of how a large group of the world's top economists think on a topic, I search here to see if that topic is covered. Most importantly, when they all lean one way or another, it's a pretty solid sign that is the best knowledge mankind has on that econ topic.

I searched for bitcoin, found four topics, and ouch! I guess my understanding of econ and BTC was not far off. Check out these [2,3,4,5]. It's worse than even I guessed.

After this much expert evidence, I don't even have to question my understanding of it much more. The crappiness of BTC except for scams and bubbles is not much of a question any more.

[1] https://en.wikipedia.org/wiki/Economics_of_bitcoin#Views_of_...

[2] http://www.igmchicago.org/surveys/bitcoins/

[3] http://www.igmchicago.org/surveys/bitcoin-ii/

[4] http://www.igmchicago.org/surveys/bitcoin/

[5] http://www.igmchicago.org/surveys/cryptocurrencies/

What would have to happen for you to change your mind ?

What if bitcoin passes gold in market cap ?

What if central banks start holding bitcoin like they hold gold rn?

What if a future Nobel winning economist says something nice about bitcoin ?

Or is it impossible for any new information to change your mind on this ?

Popularity != sanity, and popularity != value, and popularity != efficiency.

See: beanie babies, furbies, tulips, leaded gasoline, leaded paint, the rise of fascism in Europe in the 1900s and so on. A lot of people using something doesn’t make it at all valuable, efficient, effective or good — anything, really. It has to do that on its own merits.

You can’t escape the fundamental economics here. It’s bad. Rather it’s great at separating people from their money, and allowing the trafficking of illicit narcotics and money laundering.

[edit] let’s look at it the other way. Why isn’t the popularity of the dollar enough to convince you the dollar is better? Who would have to adopt the dollar to change your mind?

You are right, dollar is still better for facilitating most illegal behavior. I like the dollar.

I’ll tell you what would convince me bitcoin is bad: discovery of a bug that allows double spending retrospectively and someone corrupts the entire ledger to date. China taking over and re-educating me it’s bad. God addending the Old Testament or the Koran that it’s bad.

>What would have to happen for you to change your mind ?

It's not about changing minds - it's about evidence. There's ample evidence BTC is simply repeating the flaws of discarded money systems, and they were discarded because they are terrible. Unless BTC addresses these flaws there is little sense in painting a pig.

Mankind has learned that a money system that cannot adjust to shocks is a terrible idea. So first BTC needs this ability, which would make it not BTC. If it cannot at least meet minimum requirements, then it fails. You might as well argue we should trade tulip bulbs as currency if you're going to ignore modern evidence.

The real question is why do you clutch BTC given centuries of evidence and widespread professional opinion that it fails at the things a money system needs to be modern? Why do you insist on repeating flaws that have been conclusively demonstrated to fail?

I gave you ample phrases to google and learn from. If you still claim BTC is somehow immune from all its flaws that is your choice.

So, no Plausible scenario Of new evidence for you to change your mind then ?

Those trusted intermediaries require a lot of identity information, usually by law. It’s nice to be able to transact privately sometimes.

Iran and Venezuela would disagree, or any country the US government has banned from the international banking network. The fact that one country can unilaterally block bank transfers between third parties solidifies the need for non-politically centralized banking.

Because allowing the transfers enriched those in power, helping keep the leadership in place and the people down. For the record this is something you can vote on and change, because it’s a group decision. Unlike all those poor scammers who keep getting blacklisted from exchanges and such.

Secondarily, crypto is no panacea here. You can’t solve the initial distribution problem. How do you decide which individual gets how many coins to start?

If you ignore that issue, and if you can only trade within the country the total amount of crypto and the total amount of local currency remains the same meaning the total net welfare has not improved. If you can trade outside the country a myriad better options exist.

Crypto doesn’t help here either, it just purports to through magic of the blockchain. What’s necessary isn’t magic beans that live in computers but for the locals to rise up and overthrow their local governments and institute proper monetary and fiscal policy.

Concrete example: why don’t people use bitcoin in Venezuela (separate from the fact a single transaction costs a few weeks wages)? It’s been 11 years and were still waiting like skeletons for op to deliver. How did the Petro change the situation there?

People actually use Bitcoin in Venezuela [0], not en mass, but still, it provides some very useful escape hatches for a part of the population.

0. https://www.bbc.com/news/business-47553048

Part of the population solves none of the problem, as I outlined. The monthly minimum wage is 1 bitcoin transaction fee. It's only the wealthy that can afford to use it in the first place. It's doing nothing to increase the total welfare of the system.

I never disputed that a select few couldn't benefit from it, it's not a good general solution, though.

Then use dai? Much lower fee. When ethereum scales to thousands of transactions a second, it will be the top solution.

That does not solve the initial distribution problem.

I believe he’s saying the opposite - that people like to think of blockchain as trust in technology, but in reality you are still trusting in people.

Any evaluation of the security of the system has to take the whole socio-technical system into account. Too many blockchain enthusiasts focus on the technology and ignore the rest.

Spot on.

While I agree with Bruce assessment that a lot of Blockchain evangelists push the whole "decentralized trust" or "trustless" or "in code we trust" ideas, I actually think you don't need to buy these ideas to see some hope in how things may work "better" if the promises of Blockchain materialize:

I agree entirely that we would need still be trusting humans even with Blockchain (shifting trust around). But I do see us gaining something with Blockchain: standardization. Rather than having multi-national and multi-lingual team of lawyers and diplomats to establish contracts and rules and how to interpret it and enforce them we can (try to) standardize all in code.

We would all still need to trust our lawyer/engineers, but now they don't need to trust some intermediary, translator/etc to get things done.

People think lawyers are just about making sure there are no loop holes, but its really not that. Lots of it is about making sure that two parties really agree on something including in the fine details.

Blockchain isn't going to magically solve the problem of having to figure out what the two sides actually want.

It won't, I agree. but won't it help auditing compliance and enforcement after agreement has reached?

Maybe if everything the agreement is about is on the blockchain.

However i cant think of a non-contrived example where that is true. Real agreements involve real things in the real world, outside of the reach of automated compliance checking, and you're back at square one.

Perhaps you could say that the formalized computer language helps disagreements, but lawyers already have formalized language for this, and i suspect legaleze is better for making human-agreements than some turing complete language ever would be.

I found the value of blockchain in a low trust setting easier to understand after reading up on the Byzantine Generals Problem fault tolerant distributed systems.

> Bruce's take can be summarized like this: with a blockchain, you're choosing to trust code and cryptography instead of a person. And when immutable code does the wrong thing, you can't trust them to fix it like you could a person.

Maybe Bruce hasn't heard of the Ethernum DAO fork [0], or the two subsequent forks. The immutable block chain seals a bad decision, then suddenly it wasn't so immutable any more. Just like someone paying you via direct bank transfer is irreversible, right up the money they are transferring was stolen from the bank - then it turns out it's real easy to reverse.

Crypto currencies are the tools of people, not the other way round. If we, or in the particular case the miners decide they want the block chain to be mutable, then that is what happens.

[0] https://en.wikipedia.org/wiki/Ethereum#The_DAO_event

"the people" don't kid yourself for a second that cryptocurrencies aren't the purest form of plutarchy. Who owns all the mining power?

No one ever seems to address the fact that Blockchain still doesn't magically provide trust at all. It still needs people to participate for that and provide their resources (miners).

And for people to participate they need some incentive or reward. So how do you ever motivate enough people to participate in your distributed ledger?

You can give them blockchain tokens of course, but now you've created a new currency again with all its associated problems. Hardly a solution for your original problem of low trust.

Bruce's take can be summarized like this: with a blockchain, you're choosing to trust code and cryptography instead of a person

It can't be summarised like that. He explicitly says: "Any evaluation of the security of the system has to take the whole socio-technical system into account".

His issues revolve around the fact that blockchain ecosystems have become centralised: "Just about everyone using bitcoin has to trust one of the few available wallets and use one of the few available exchanges. People have to trust the software and the operating systems and the computers everything is running on. And we've seen attacks against wallets and exchanges. We’ve seen Trojans and phishing and password guessing. Criminals have even used flaws in the system that people use to repair their cell phones to steal bitcoin".

This is why the Stellar blockchain looks promising. On the Stellar blockchain, you put your trust in anchors and issuers (human entities) and each blockchain-subset (roughly the trust-union of specific-token issuers and specific-token anchors you choose) can create a highly distributed / decentralized blockchain from each users perspective, depending on how many anchors and issuers you are willing to trust.

It's a wonderful hybrid of "high-social-trust" networks run by humans (banks, governments) and distributed trust via code / your choice.

> with a blockchain, you're choosing to trust code and cryptography instead of a person. And when immutable code does the wrong thing, you can't trust them to fix it like you could a person.

There are soft and hard forks that do change the behavior of blockchains so I'm not sure I entirely agree with the premise of this argument. Help me out, what am I missing?

So far I haven’t seen a usecase that absolutely needs a blockchain either. I mean one that can only be addressed by a blockchain and will fall apart without it. But many people have used it as a general enough hammer to hit every nail they can find with it. As last it as people make money on it, even hypothetical money, it’s going to stay popular.

I think that making money part is due to a game of musical chairs. No actual value is generated.

> No actual value is generated.

I actually (unfortunately) don't think this is true in a financial sense.

The only people with an incentive to use a semi-anonymous "currency" with extremely high transactions fees are criminals. You can argue that not all crimes are harmful or immoral, and you'd be right, but that's outside the scope of my comment.

The more crypto is used as a payment for crime, the more value it has. It becomes an investment vehicle for crime. If you think global crime will increase (and increasingly use crypto), then you can invest in it using Monero or whatever people think offers the best combination of liquidity and privacy these days.

I think you are right in a narrow sense. But from the perspective of a society, it seems to create 'negative' value.

I have a few words to say on that.

https://louwrentius.com/cryptocurrencies-are-detrimental-to-...

Facilitating transactions is a valuable service.

That tech already exists with a fraction of the transactional cost, in the far majority of situations.

That's a different claim than you started out with. Additionally, remittence gets a lot harder internationally. Some of those services charge very hefty fees.

Being the best at something is different than not doing something at all. You don't have to be the very best at something to generate value doing it.

Shitcoins can be fun. Fun can be valuable.

No, when the flow of money is constrained it interupts actual labor. The great depression is the classic instance where if money had been made work would have been done and everyone would have been better off.

What does "generating value" mean in this context. Aren't most transactions just shifting resources around?

If i have a bunch of cookies and my friend has a bunch of cup-cakes, and i trade some cookies for some cupcakes, what abstract value was created? The total number of cupcakes and cookies stayed the same, they just shifted around (like musical chairs). Neither me nor my friend are any richer, although we may be happier. Was this transaction useless?

Buying drugs/murders on the darknet and getting untraceable ransoms from your cryptolocker malware are at least two solid use cases. One could argue evading currency controls is a 3rd.

The problem is that there's been no new uses found for blockchain after the initial uses discovered.

Blockchain isn't completely anonymous though. Bitcoin, for instance, can be and has been deanonymized to catch criminals. The entire concept of a blockchain is create an extended ledger, which isn't exactly an idea focused on privacy.

It's an idea focussed on.... what exactly?

It seems like a solution in search for an actual problem.

That's not an inherit limitation of a blockchain. Monero exists. It actually delivers on privacy where bitcoin couldn't.

Monero is private and has taken over as the far better option for the few real use-cases

In that vein, I really like this book:

http://www.americankingpin.com/

About the rise and fall of The Silk Road

To those taking this comment as a joke - it is not a joke.

Regardless of the morality/legality of a use, the fact that those are real use cases for the currency do give actual value to bitcoin as a currency.

Now it's worth noting that "digital gold (speculation)" seems to far surpass ransom+online drugs as the reason for bitcoin to have value.

> Now it's worth noting that "digital gold (speculation)" seems to far surpass ransom+online drugs as the reason for bitcoin to have value.

That's not really value, nothing is created. It only seems a game of musical chairs, with some ending up with 'lambo money' and the rest just lost money.

Terms like ponzi scheme or piramid scheme (it seems to contain elements of both) do apply.

> Terms like ponzi scheme or piramid scheme (it seems to contain elements of both) do apply.

How so? Bitcoin doesn't pay people out of the proceeds of other people buying in. Its value is solely a function of demand and scarcity.

Sure compare it to beanie babies, tulips or gold. Speculation may be stupid, and it may very well end up with most people losing their money when the bubble collapses. But it doesn't have very much in common with pyramid schemes or ponzi schemes.

Most cryptocurrency investment schemes basically work like this. Someone creates a cryptocurrency. The creators premine the cryptocurrency and sell the currency at a fixed price for pennies. The early investors try to "shill" their cryptocurrency and market it to as many people as possible. There is an influx of cash and the currency is valued at a few dollars. Early investors sell and late investors sit on a worthless cryptocurrency and the only way for them earn their money back is by marketing the cryptocurrency to even more people and selling it to them. Rinse and repeat.

Sure, i agree that many of the non-bitcoin altcoins are pump-and-dump schemes.

Although lots of financial fraud has similar elements, this seems very much not a ponzi scheme (users arent paid with other users investment and not a pyramid scheme (users aren't explicitly encouraged to recruit new people nor do they get an explicit cut of newbies). Pump and dump seems like an appropriate description of what you are describing.

I also don't think that applies to bitcoin, which was the cryptocurrency we were talking about.

It's the other way around. Bitcoin pays miners who keep the network running. It does so by creating bitcoins. Its not peoples money that is used to pay them its indirectly trough inflation. No one cares about that because price gains have been way higher than inflation loses. Also the upper hard cap of 21MM makes people think inflation doesn't exist. Yet it of course exists until the block reward is actually zero. In other words the price has to go up forever doubling at lest every 4 years on average to cover its own running cost with the simultaneously halving reward. It's obviously a scheme that will collapse, it's just designed to possibly last many decades.

I think those are all basically the same use case - they're all about routing money of nontrivial value around governments. Much like Tor, it is not 100% unblockable, but it is quite difficult to block if you keep the internet flowing at all.

And like Tor, it can be used for both moral good and moral bad. (In fairness, you can decide that the good outweighs the bad for Tor but not for cryptocurrency; you can also believe that the good doesn't outweigh other harms like the computational/energy cost of running a secure blockchain, which necessarily needs a nontrivial fraction of the world's purchaseable computing power.)

Number 4: Having a highly volatile, non regulated investment option with a low barrier of entry that is probably controlled by a handful of bitcoin heavyweights.

A blockchain decentralizes trust, you can always solve the same problem a blockchain tackles by introducing a central point of trust. The question is if this choice is acceptable, more resilient, cheaper...

I think the case is made by Bruce that the whole concept of decentralised trust is an oxymoron.

In the end you do trust the code, a computer running the code or some currency exchange.

You trust a decentralized network of computers running code, not just a single one, that's the main point.

There's both centralized and decentralized exchanges available.

A bank is not a single entity, it's made of people converging after reaching consensus

There are a lot of banks, with different interests that usually compete with each other and work under regulations that have been discussed and approved after reaching consensus

There are several different central banks, that work together under regulations discussed and approved because they reached consensus

The "decentralised" exchange already exists

So far I haven’t seen a usecase that absolutely needs an Array either. I mean one that can only be addressed by an Array and will fall apart without it. But many people have used it as a general enough hammer to hit every nail they can find with it. As last it as people make money on it, even hypothetical money, it’s going to stay popular.

Here's a usecase:

Blockchain can be used to issue something like Apple stocks in a decentralized manner which achieves million times more scale and lower cost than what's possible now.

Obviously a centralized entity like New York Stock Exchange could also copy the technology of using Merkle hashes and possibly achieve much bigger scale, at the end of the day it would still be the coordination issue for a single organization.

At the end of the day blockchain solves the coordination problem in a low trust environment in a definitive manner. As it turns out that the cost required to solve this coordination problem is really high when you compare it to the high trust, but low scale environment. but blockchain manages to achieve planetary scale throughput (obviously not right now because this is the foundation layer, things need to be built up).

It's like cash versus Visa. Cash is possibly billion times more scalable than Visa because Visa needs to reconcile a single database but cash transactions do not, as long as the recipients locally verify that they're receiving a real bill.

This article is from 2019 but I discovered it only recently by accident.

Bruce Schneier makes the case against both cryptocurrencies and Blockchain technology, as I understand it.

I think it should get some more attention, because it makes a very good case, centered around the topic of 'trust'.

I noticed that this article has been discussed on HN a while back:

https://news.ycombinator.com/item?id=19097613

Ah, I did not notice that the article was written by Bruce Schneier. I need re-read the article now. Weird how trust works.

Store your private key on a paper wallet all you want, but what are you going to do during a tiger kidnapping? People look at me weird when I bring this up, but know nothing about real security. I could advice some ISO certified procedures to keep your money safe but your family is going to be the weak link. Go read In Cold Blood by Truman Capote to get an idea on what could go wrong.

A large chunk of the arguments tend to assume Proof of Work as the consensus mechanism which isn't a flaw inherent in blockchain (eg. could have Proof of Stake, etc.)

I think the point is more that whatever tech you use, you eventually still have to trust humans or an organisation.

There are always inherent risks that you just can't reduce with tech.

Yeah but if I can trust them in 10 seconds with very little effort on my part, I can be much more nimble with my strategy. If i can quickly verify the validity and source of primary data, I can make decisions on that data without hesitation. The value is in automating the trust, having a tiny little notary saying yes this is true, go ahead and make evidenced based decisions. Now of course we could do this with other tech, but I find the standardization of this trust mechanism to be the substructure of new business building tools and services around this quick trust, much faster than they can build them around trust accumulated by institutions, governments, notable individuals.

Are there any good examples of working PoS?

Tezos

I've not seen any satisfactory explanation for how a working PoS system would operate. Granted I'm not an expert, but anytime I delve into whitepapers, they are written confusingly and without crucial details, which makes me wonder if some of these alternate chains are just vaporware or at the very least a lot less than they promise.

This is a misunderstanding of the concept of "trusted third parties".

Incredibly, yes, you have to rely on the software that you use to perform tasks to perform them. But this is not the same as placing another person in a mediating position where they can choose to betray you.

Confusing these two senses of "trust" results in an intuitively appealing but entirely mistaken family of arguments that aren't the gotchas people think they are.

The Very Short Introduction book on Trust covers different senses of the concept of "trust", including the ones I mention here, in a readable and compact way that I highly recommend.

> Yes, bitcoin eliminates certain trusted intermediaries that are inherent in other payment systems like credit cards. But you still have to trust bitcoin—and everything about it.

.... which is very tolerable to me.

Right now, it is more expensive to use blockchains do to both the novelty and specialization of it. As that becomes more ubiquitous it doesn't.

Whether we are talking about FX spreads to and from fiat currencies, or onchain transaction fees due to congestion, these all get better due to pressure and competition. A reason to improve. A permissionless ability to develop the improvements.

Even in things people pretend not to care about (after they lose money) like token sales, it costs token sale issuers more money to list on a random Asian exchange, than it costs to list on the Nasdaq. When the permissionless nature of the technology should mean the opposite is true. More competition and it fulfills that.

For how long will blockchains be considered a novelty? We are 1 decade after bitcoin release.

Enterprises are just moving away from their private blockchain to public blockchains, pretty much just this year, following Ernst & Young's Baseline framework.

Outside of the private sector, the regulatory frameworks specifically for blockchain technology have only begun to been ratified over the last 18 months, with the key market under the US Federal Government just getting providing clarity to financial institutions last week, at the Office of the Comptroller.

The metric you are referring to is a misattributed S-Curve of adoption that has condensed for a few select technologies over the last century.

Bitcoin is not the primary area of development for industry use of blockchains. There are various virtual machine technologies proliferating that bitcoin does not feature.

I'm not technical enough to disagree with Bruce. I'm also not an international law lawyer, fixer, intermediary, etc, to agree or disagree with how we have been trusting intermediaries for thousand of years and why this is impossible to replace.

The way I see it is that blockchains and smartcontracts have the potential to replace treaties, lawyers, corporate intermediaries, etc in an increasingly multipolar world where treaties would serve as a legal protocol to regulate transactions.

Maybe it is naive (I'm still trying to workout how this can ever happen if nation-states actively oppose these desintermediations) but I do think that privately orgs and people could benefit from technology that helps them transact with less middlemen.

In a world where China/Russia/etc is trying to avoid exposure to US money/laws maybe this could serve as alternative?

Bitcoin is really about power and control, not about trust. In many aspects it's a rebellion of coders against the MBAs and lawyers.

China has been trying to reduce dependency on USD. They would probably wish that the world use Reminbi to deal with them but world won't trust China's money (I also understand that China way of regulating it's currency also prevents it from propagating out of China) so, how about China adopting some neutral stablecoin like DAI for certain Dealings?

I think about Bitcoin (and similar systems) primarily as a long term store of value.

I think they are a pretty fundamental innovation in that space.

What are good alternatives? Gold, cash, or ETFs have substantial drawbacks - especially in situations where property rights are insecure and inflation runs wild (or just eats up savings at a somewhat moderate pace, as is the case in many developing countries).

Of course Bitcoin is also a human system, eventually, but what you are trusting is an emergent, non-hierarchical order which should be more resilient than hierarchical systems. The technology facilitates the emergence of such a system.

I actually think the flawlessness of the technology is not a necessary condition of Bitcoin's success. If bugs are found then consensus in the social system can emerge which can minimise damages

All of that being said I think the use case for blockchain is probably quite narrow.

> All of that being said I think the use case for blockchain is probably quite narrow.

Name one, I dare you.

Name something that really matters. That will offset all the damage they do.

The use case I mention in the first sentence of my comment: Long term store of value.

If that's the case, then it does a terrible job at it. Cryptocurrencies, by and large, are absolutely horrible for a reliable store of value. The swings in value are too great and numerous to count, based on the whims of whatever people feel like they are, often on a daily basis.

I'll take US Treasury notes any day for "long term store of value". If the United States gets to the point where they can't honor Treasury notes, we will have much bigger problems on our hands anyway.

See the "long-term" part. On most timescales since its inception, btc has stored value pretty well.

I totally read that part. It still sucks for it because of how unpredictable cryptocurrencies are. Bitcoin has a bunch of "Bart Simpson" style pump and dumps and, even over the course of the many years of Bitcoin, unless you were a true believer from the early days, I'd find it hard to believe you came out ahead or didn't lose your shirt in the process. I'd rather not be subjected to that, and there is absolutely no guarantee bitcoin will be relevant 10 or 20 years down the road. Another black swan event, a cryptocurrency that becomes the new hotness, or maybe even a quantum computer breaking into wallets could render bitcoin worthless.

Another thing to consider, most users of bitcoin or whatever else generally relies on exchanges to get out what they are really after, US Dollars. The end game is to see which sucker will buy your tokens for more than you paid for it, then you can cash out. But to that end, we kinda have to trust the exchanges as well, and there are many cases of fraud that have happened at that level, which shakes trust in such a system.

Now if we are being intellectually honest here and asking ourselves if bitcoin and its ilk is good for something like speculation or even as the basis for treating it as an "exchange" for transferring money (albeit a rather low performing one at that compared to other established systems like SWIFT), then I can potentially see the benefit in that, even if I personally have no real use case for it.

It seems like a really bad candidate for a long term store of value. For one, just look at the price volatility - that is the last thing you want in a value store.

Second, the fact that it’s ONLY use is as a long term store of value is red flag for being a GOOD long term store of value. If there is an adverse economic situation (which is one of the primary times you are going to want to tap into that long term source of value), the price is going to collapse... because everyone will be needing to pull OUT of the store, and no one will be wanting to buy.

Funding niche software projects by using tokens. For example, Unitrade launched a token to fund development of a project providing advanced tools for Uniswap, a decentralized exchange https://unitrade.app/ https://uniswap.org/

Decentralized autonomous organizations (DAOs) that are new and experimental forms of company formation. See DXdao https://dxdao.eth.link/#/

Compound, a large player in the Decentralized Finance (DeFi) space that allows anyone to borrow against their crypto holdings https://compound.finance/

The craziest part of Hacker News is the near universal opposition to cryptocurrency and blockchain technology. It is not going away, and governments are finally bending to reality and accepting it, rather than banning Bitcoin and similar outright.

There is no better way that a developer can earn outsized riches than blockchain and cryptocurrency. You are solving tough problems in difficult, constrained environments and are rewarded extremely well for doing so by a global group of investors involved in this alternative financial system.

But please, if you don't like crypto, just build whatever SaaS bullshit an MBA is telling you to and take your 0.1% equity for busting your ass. I have spent years in the extraordinarily interesting world of blockchain and have been rewarded extremely well in terms of personal growth, education, and financial rewards.

> There is no better way that a developer can earn outsized riches than blockchain and cryptocurrency. You are solving tough problems in difficult, constrained environments and are rewarded extremely well for doing so by a global group of investors involved in this alternative financial system.

You are not rewarded for "solving tough problems" or anything else related to creating actual value. It's FOMO driven gambling, scams and enabling organized crime that gets money into the crypto ecosystem.

> But please, if you don't like crypto, just build whatever SaaS bullshit an MBA is telling you to and take your 0.1% equity for busting your ass. I have spent years in the extraordinarily interesting world of blockchain and have been rewarded extremely well in terms of personal growth, education, and financial rewards.

> What are good alternatives? Gold, cash, or ETFs have substantial drawbacks

The search for a good, long-term store of value has been around since the invention of money. Adam Smith's "Wealth of Nations" discusses this problem since the beginning of capitalism.

And, BTW, Bitcoin is no better. It's only worth what people are willing to pay for it. Just ask anyone who bought when it was worth around $20k! The only way they'll get their money out is if another sucker comes along.

A good store of value's value does not fluctuate.

The point of investments like ETFs, mutual funds, stock, ect, is that you're allowing someone to put your "stored value" to use, so they can create more value for you.

But, here's the thing: If society crashes, it doesn't matter what you've put your money in, we're all screwed.

Are blockchains truly decentralized when a small central group of developers control their codebase? Sure, you can choose not to use the latest version, but then you're likely not going to be part of the main network.

Interesting what would happen if FBI (silently!) takes over the github bitcoin repository and starts pumping out a version spying on all transactions.

Transactions are already public.

names, addresses and SSNs of transacting parties are not public

You will not be on the main chain if the modified version changes some consensus rule. That's called hard-fork. Look at the segwit2x debate to see how this worked in real world.

If a modified version doesn't follow the same rules, it will never finish the initial sync.

And you can still sync with very old clients. Most of the work on the bitcoin client is optimizing validation. Consensus rules are rarely touched.

blockchain - the wrong answer to any problem

Indeed, blockchain: an answer to a question asked by nobody.

but the USPS just filed a patent for a voting system based on one: https://heraldsheets.com/us-postal-service-usps-files-patent...

Well that nullifies all Bruce's arguments, right?

Not if the transactions can be deanonymized as it has happened to Bitcoin.

That didn't 'happen' to Bitcoin. Bitcoin was designed that way.

Sorry if I gave that impression (enough to warrant negative votes), but that was not the point I was trying to make. Just offering a contrasting link.

I work broadly in financial markets. BC was talked into the ground thankfully ending 2018. Haven't heard squat since.

I was genuinely interested particularly on aspects of dis-intermediating custody banks. But it doesn't take long to realize a few things about bonafide public BC:

- the random guy adding his/her node to the network will need to catch up on serious data amounts or just take somebody's snapshot "as word". Can you boot strap your new node with a snapshot. Should you? Well, what are you going to do about getting all the back data in?

- nobody is going to put private financial/trading/whatever details plain text on a block. so it'll have to be encrypted. so what are we really verifying? you're verifying what my copy of PGP produces? whatever gibberish I post to the block? I don't want to disclose from/who/when/amount/shares/bank-info/exchange/stock symbols .... nothing ... but regulators may need to see data un-encrypted ... so now what? Send regulators data the old way? While I do CAPEX the new way? C'mon.

- The data has to be represented somehow. Since BC brings nothing to that party ... BC will wrap Swift, FIX, Omego, or by exchange or by regulator reporting format ... so BC is a crap load of work to fix nothing about the domain. And even that is a simplification. There's lots of little gotach's in data reporting between all the counter parties.

- throughput is an issue on BC

- just the idea that any joe can add a box to the network and start looking at financial data ... it's a non-starter ... although we can argue you can get some visibility from exchange data ... but traders might use BC to help settle and that may require a heck of a lot more detail that what NASDAQ provides

- BC isn't a currency. A currency has to do certain things including being universally accepted in the country, have near constant value day to day. Whatever that is BC doesn't do it.

- Another thing I find jaw dropping about the BC mentality is the insistence the unshakable foundation of the blockchain insofar as its non-deprecation and on the computers to apply the same logic and/or verification and xfer fees without human interference or bias. To date BC is pre-pre-pre-neophyte status. You start bringing in serious economic pressures, 1-offs, crashes, bailouts, disagreements and I'll bet some or all that code will fail/falter in human hands leading to code changes leading to forking. Today's money system didn't start this way. It was beat into today's shapes through problems, mistakes, and here and there some good ideas. So will BC. I seriously doubt the immutability of the rules governing blocks.

Now you can back off public, decentralized ... into private block chains with shared or not shared keys and the like ... or just point-to-point block chains between trusted and distinguished entities ... but very quickly ... so what are we really doing here?

BC is used it seems to me for speculation, buying small potato things like coffee or for the coolness factor or hiding. I'm not seeing serious industrial use cases.

Now, there are two areas that I might see BC useful for:

- things like ownership, proof of (say houses) - maybe voting

All that and you still think housing records should be on a public blockchain? What happens if someone steals my private key? Do they now own my house? If we rely on public records and the legal system to sort it out then why have a decentralized and immutable database in the first place?

> All that and you still think housing records should be on a public blockchain? What happens if someone steals my private key?

Then you lose your house.

Here's the alternative: someone steals your house, and they didn't have to steal your private key first. It happens every few years. Usually the victims have been overseas for a while.

And there are many good reasons not to trust Blockchain evangelists.

I just lost some respect for Bruce Schneier.

Why? Could you elaborate further?

But do you have any decent counter-arguments against his case?

He does talk about the standardization aspect as a plus to blockchain, the question is, do we want a bad standard now or a better standard never?

But do you trust him?

Slow clap.

Yeah, it's called confirmation bias. Many people heavily invested in fads exhibit it.