Hacker News new | comments | ask | show | jobs | submit login
Blockchain and Trust (schneier.com)
116 points by bb88 5 days ago | hide | past | web | favorite | 107 comments





Cryptocurrencies doesn't remove all trust. It's about removing trust in third parties to send and receive transactions. That's it. And that's a big deal.

> Honestly, cryptocurrencies are useless.

Honestly, he makes some good points but this is ignorant.

For example:

* Uncensorable donations to wikileaks

* Avoid having PayPal or other payment processors freeze your merchant account for X months, on a whim

* Avoid the risk of charge back fraud for online payments

* Ability for weed shops and porn sites to accept digital payments

* Fleeing Venezuela with your wealth intact

* Send money to and from countries where no banks or western union operate.

* Financial privacy for everyone not just the ultra rich (lookup Monero before commenting)

Cryptocurrencies have some very good features.

There are issues of course, but how can you claim all these are useless?

> For bitcoin, that's the cost of the additional bitcoin mined

No, that's the cost of securing the network to prevent double spends.


I can think of counterpoints:

> * Uncensorable donations to wikileaks

Untraceable theft from your wallet

> * Avoid having PayPal or other payment processors freeze your merchant account for X months, on a whim

Unregulated merchants who aren't incentivized to play by the rules

> * Avoid the risk of charge back fraud for online payments

No third-party intermediary in the event you get ripped off.

> * Ability for weed shops and porn sites to accept digital payments

Ability for weed shops and porn sites to accept digital payments (This depends on your stance on these issues. Not everyone feels the same as you).

> * Fleeing Venezuela with your wealth intact

Fair enough

> * Send money to and from countries where no banks or western union operate.

Indeed this is useful, but seems like a small use case relative to the hype around the tech

> * Financial privacy for everyone not just the ultra rich (lookup Monero before commenting)

If you buy something on ebay, ebay and the credit card company have a record of your purchase. If you buy something from a bitcoin merchant, they will have a record of your purchase. How is that different?


Some counterpoints:

> Untraceable theft from your wallet

Blockchain analysis is a flourishing field. Vanilla bitcoin is far from untraceable, in fact it is LE’s wet dream.

> Unregulated merchants who aren't incentivized to play by the rules

There are two distinct sets of rules here. The ones merchants have to legally abide by (government imposed) and the ones that fit PayPal’s agenda. The latter is usually the problem (c.f. WikiLeaks blockade) because it is entirely based on the whims of a private company.

> No third-party intermediary in the event you get ripped off.

I often see this argument. Schneier also mentions it as a drawback. Note that reversible transactions can be built on top of bitcoin, if it is desirable. The opposite is not true. At least bitcoin gives you the option to do irreversible transactions if you wish.

> Ability for weed shops and porn sites to accept digital payments (This depends on your stance on these issues. Not everyone feels the same as you).

People need to be able to break the law and operate gray markets to move society forward. See Moxie’s excellent essay on the subject [0]

> If you buy something on ebay, ebay and the credit card company have a record of your purchase. If you buy something from a bitcoin merchant, they will have a record of your purchase. How is that different?

The reason eBay needs your real name is to combat chargeback fraud and tax evasion (for merchants). The former is irrelevant with irreversible transactions, the latter remains valid in that particular case. But what about Joe-random-SaaS? Why does {Spotify|Shopify|Expensify} need my real name besides payment?

[0] https://moxie.org/blog/we-should-all-have-something-to-hide/


> Untraceable theft from your wallet

Theft is traceable (by looking at the blockchain). Theft is untraceable in the fiat system, because you don't have access to the bank's coffers to check they've not pilfered your money. Turns out they never had the coffers anyway, your money is just some number on a screen, and there are people who have the ability to change those number arbitrarily.

If you take the right steps to secure your bitcoin private key, it doesn't even need to exist in digital form, or physical form, ever. It can be purely inside your head. It's just a number. One which the probability of somebody guessing correctly is unimaginably small.

> Unregulated merchants who aren't incentivized to play by the rules

The same is true for fiat money. Scammers exist, and if you get scammed, you will usually be stuck for months without access to your funds while some third party performs an "investigation" (In paypal's case, "investigation" means return money to the buyer).

How about unregulated "customers" who are not incentivized to play by the rules? Credit card fraud is a huge cost to businesses.

> No third-party intermediary in the event you get ripped off.

The correct term is choice of third-party intermediary (or not). Bitcoin does not force an escrow on you, but allows you to choose whether or not to use one, and who to use. This process can be conducted digitally with multisig transactiton outputs.

> (This depends on your stance with these issues. Not everyone feels the same as you).

Feels are overrated. What are you going to do about it? We've already seen that action is worse than inaction in drug wars. If you think that paying for porn is immoral, what about the morality of you declaring unilaterally that you are cutting off the income of women because you don't like them getting naked in front of a camera?


Sure it is possible in Bitcoin. But Monero is frequently used as prime example of an untraceable cryptocurrency. The reason is because of Monero's technology. But in the same vein any stolen Monero cannot be traced.

> Theft is traceable (by looking at the blockchain).

So the people who stole the money from Mt. Gox are currently sitting in jail, right?

This also shows the doublethink you sometimes see with cryptocurrency: it's totally anonymous, and so you can use it for privacy, except for when it's not, so you can use the immutable record to figure out who harmed you.


I was referring to the ability to discover that your funds have been stolen, not necessarily who stole them. (The latter usually doesn't matter, you probably won't see the money again). The fiat system has no ability to monitor for theft. Theft can go on for years unnoticed, and by the time it is noticed, the entitiy involved may be insolvent, having a bigger effect on many people, or even the larger society if the entities are big enough.

Bitcoin is not totally anonymous. If you trade through AML compliant entities, your details will be linked to transactions, and people might be able to discover who you transact with based on the public transaction information. With coinjoin and similar fungibility enhancements, that process of tracking is made far more difficult towards impossible. Also, the Lightning Network enables anonymous payments through an onion routed network, where the only public information is who you have open channels with. Anonymity will become the norm in Bitcoin even if it not the case yet.

If you look at it from the perspective of a merchant, rather than from a "consumer protection" point of view, then a merchant generally does not know who defrauded them with stolen credit cards. In general, it is rarely worth investigating and businesses will generally count their losses. Entities like Paypal are notoriously bad for the seller as they side with the buyer most of the time - it isn't their money so they don't need to take it seriously.

Bitcoin will benefit merchants, who can take payments which are irreversible, except by their own decisions for refunding. Consumers will need to take a bit more responsibility to vet the reputations of the merchants they deal with, instead of always being bailed out for their poor decisions.


Yes they are fair counterpoints.

> If you buy something from a bitcoin merchant, they will have a record of your purchase. How is that different?

Sharing your data with multiple merchants is much better than allowing one or two entities track all your digital payments, like your credit card provider and your bank.


Regarding your last point: it’s different because it’s relatively trivial to create as many wallets as you want and or use a form of ring sharing. That allows an individual to express financial visibility and transparency in the way they see fit.

Thank you. Assuming you buy something that requires shipping, there would still be some sort of paper trail though, right? If so, this assertion would only stand when the goods were digital or counterparties were physically present and wearing masks.

Even if it was physical you wouldn’t tie the financial picture to the user. The wallet I purchase the physical good with tied to my address may only show $50 worth of coin for example. Compared to my entire net worth.

You also to have to ensure that the two wallets are not connected to each other by other means, such as feeding them with currency from the same exchange, or revealing the same IP with them to the same merchant, or the like.

This is trivial with zksnarks. And that looks like some form of that idea will become standardized.

> It's about removing trust in third parties to send and receive transactions. That's it. And that's a big deal.

Uh huh, so when in 2017 I wanted to send my bitcoins to exchanges to sell them, I could not, since the miners did not increase the block-size.

How is that for removing trust? This happened to me thrice during 2017, bitcoins were not spendable.

In particular around May 2017, I needed to pay for a VPS, a small amount about €5 worth. The transaction fee of bitcoin to get the transaction processed would have been more valued than the "pay for VPS".

Useless is a compliment.


One only needs a single instance where Bitcoin worked when other payment processors did not. Wikileaks fulfills that example. Sci-hub is one more example than I need.

Piling on a thousand other obvious cases where cryptocurrencies are useless doesn't change those edge cases.

Just to get another example where this logic plays out: I'd bet the vast majority of cases where people are using GPG, the feature of "irrefutability" isn't needed. There are probably also cases where the user doesn't realize that is a feature, as well as cases where it's dangerous to have that feature. But I'm not prepared to call that feature useless, and if I claimed that it was it would only take a single edge-case from a contrarian to prove me wrong.


This is a straw man. Just because transaction fees were too high for your transaction to be economically viable, does not discredit the fact that bitcoin removes trust in third parties to send and receive currencies.

It is not up to the miners to increase the block size, in fact miners were very much in favor of doing it. Bitcoin has other important constituents (users, core devs, wallet devs, exchanges) that would need to be on board for any modifications made to the core protocol to go through.


They were still spendable, you just had to wait longer for transactions to go through at the smaller fee. At any rate, Lightning network, the tier 2 solution for scalability in Bitcoin, has improved dramatically in the past year. There's also alternative crypto, like Ethereum, that never had such a back log of transactions.

I mean obviously you have to trust that the Bitcoin network works as it should.

And you're right, Bitcoin is broken. It's only an implementation of a cryptocurrency though and using it to dismiss cryptocurrencies in general isn't right.

See for example Bitcoin Cash (BCH) which isn't artificially crippled like Bitcoin (BTC).


> Cryptocurrencies doesn't remove all trust. It's about removing trust in third parties to send and receive transactions. That's it. And that's a big deal.

I'd like to add the trust in the issuance of the currency. A change in the issuance of Bitcoin will be an incompatible hard fork. For fiat, you trust the decisions of central banks and oversight of private banks.

Fractional reserve banking could very well be coming to Bitcoin (and we might already be seeing this to some extent on exchanges), but at least it would be opt-in.


Good point.

> Cryptocurrencies doesn't remove all trust. It's about removing trust in third parties to send and receive transactions. That's it. And that's a big deal.

You still need to trust a wallet, nodes, and miners to perform a transaction.

While attacks on nodes and miners have been rare so far, wallets have been a weak link in the trust chain. Some are malware, some have vulnerabilities, some are targeted by malware, some just have such terrible UX it's difficult to perform your intended transaction correctly.

Even if the transaction is performed on the blockchain correctly, unless it's a donation or governed purely by a smart contract, you need to trust the other party to perform the appropriate action based on your transaction (eg mail you goods)


These are problems rather than mysteries in comparison to the intrusion of "trusted third parties" as intermediaries in transactions.

Which is to say they are nice problems to be able to have.


Fleeing Venezuela with your wealth intact

Do you have real world examples of this? The only ones I’ve found involve literal fistfuls of American dollars, not crypto. I remember a play on the “crypto for Venezuela” argument was that it worked as a better currency than the Venezuelan currency, but of course it turned out that they were just using dollars then too.


I'm not sure if you realize how dangerous it is to carry around cash in Venezuela. It has an extremely high crime rate and in many cases carrying around that much cash would simply get you killed.

It is very hard to get USD in Venezuela. Are there black markets for USD? Of course, but you have to go through all the shady dealers you can imagine to get it.


Who are you going to sell Venezualean currency to in exchange for bitcoin?


> Honestly, cryptocurrencies are useless.

Cryptocurrencies are essential. I don't see this mentioned often enough: We've had distributed systems for a long while before Bitcoin showed up. One of the most significant problems with any distributed system is incentivizing people to run nodes. Cryptocurrency solved this by providing a way to give value back to people who support the network.


> Cryptocurrencies doesn't remove all trust. It's about removing trust in third parties to send and receive transactions. That's it. And that's a big deal.

I'm not sure what you mean by this. In a "normal" system, you don't trust third parties to create transactions to begin with. Third parties use a centralized system to create the transactions and you trust that centralized system.

The fundamental premise of cryptocurrencies is decentralization, although in practice it turns out to not really work that well. Most uses of blockchains end up not actually being decentralized, or not gaining any benefit from being decentralized (and definitely being less efficient than a centralized system). And you end up with situations like the DAO hack where this supposedly-decentralized system still ended up with a centralized governing body deciding to roll it back (and yeah, it was forked because some people disagreed with this, and it's kinda neat that that was even possible, but Etherium Classic currently has a market cap worth 3% of Etherium so it doesn't seem to have been very effective).

> Uncensorable donations to wikileaks

This is one of the few actual genuine use-cases for cryptocurrencies, though even here it doesn't actually work that well because most cryptocurrencies that claimed to be untraceable are in fact traceable. If you don't care about tracing, well ok, but unless you're living in an all-cryptocurrency world (which pretty much nobody is) you still have to convert to/from fiat, and that conversion point is where governments can apply pressure if they decide to act.

Though maybe one of the rare cryptocurrencies that seem to actually have genuine untraceable transactions might become popular, who knows?

> Avoid having PayPal or other payment processors freeze your merchant account for X months, on a whim

This didn't go away, just replace PayPal with "cryptocurrency exchanges". Maybe they're not freezing your merchant account but they're causing other problems.

> Avoid the risk of charge back fraud for online payments

At the cost of being unable to initiate a chargeback if you're the buyer and got defrauded. The chargeback mechanism exists for a reason.

> Financial privacy for everyone not just the ultra rich (lookup Monero before commenting)

This really only applies if you're actively trying to lie to the government (i.e. if you're doing illegal transactions and don't want them to know). Using cryptocurrency doesn't absolve you of the need to pay taxes, so for most people the idea of "financial privacy" doesn't really matter because they have to tell the government anyway.


> In a "normal" system, you don't trust third parties to create transactions to begin with. Third parties use a centralized system to create the transactions and you trust that centralized system.

For example as a merchant you don't have to trust PayPal with your money and trust them to process credit cards from the buyer and then send it to you.

You can just accept payments digitally directly like if they gave you cash in person.

> This really only applies if you're actively trying to lie to the government

I value my privacy and I don't want my transaction history to be sold to advertisement companies and the like.


> For example as a merchant you don't have to trust PayPal with your money and trust them to process credit cards from the buyer and then send it to you.

> You can just accept payments digitally directly like if they gave you cash in person.

You're just moving your trust around. You have to trust that the cryptocurrency system was implemented correctly and without bugs, and you have to trust your wallet to be implemented correctly and without bugs, and if it's a hosted wallet you have to trust the hosting service, and if it's a local wallet you have to trust your local system to ensure you don't lose your key or get hacked. Also since we're dealing with a different currency now you have to trust that the currency will actually act as a stable store of value. Or if you use something like Coinbase to accept bitcoin but immediately turn that back into USD, you're now trusting Coinbase to process these transactions.


> I value my privacy and I don't want my transaction history to be sold to advertisement companies and the like.

Merchants can sell your transaction history in any case, regardless of how you pay them.


Giving a single entity like your bank, credit card provider or PayPal your whole transaction history is quite different from giving many merchants a small part each.

It also allows you to pay digitally in person and avoid transaction records in some cases.


Cash has fulfilled this role just fine.

Not digitally.

Even if you use monero to buy a widget online they still need to ship it to you. How are you keeping your transaction history secret?

He has a number of good points. He's applied his model of trust in society, which includes four major incarnations of trust (morals, reputation, institutions, and security systems), to the topic of blockchain. This is a useful way of analyzing the utility of blockchain.

However, the article doesn't touch at all on the historical situations in which people decide to reformulate the different aspects of that model. For example, why did the French people decide in the late 1700s to rise up and recast the existing institutions of trust (i.e. the monarchy)? Because they weren't working. Similarly, the bitcoin whitepaper appeared in many ways as a response to the 2008 financial crisis. During that period, it was very clearly demonstrated that existing trust institutions weren't working and that they also wouldn't be held properly accountable.

Blockchain tech does not eliminate the need for trust. Smart people working and living in the blockchain space already understand this. In a sense, it's more appropriate to look at blockchain as a next generation trust institution (and I mean institution in a broader sense e.g. the "institution of marriage") which is more autonomous than what has existed previously. Just because the average modern day human doesn't understand and, therefore, trust blockchain technology doesn't mean that there aren't people who do both trust and understand the technology.


> Honestly, cryptocurrencies are useless.

It's always strange when people are overly specific.

Is it private money that will never find any use anymore, or is it just that they can never be digital?

The first is not an unreasonable opinion to have, given modern history and the relative success of fiscal policy (at least in the latter half of the 20th century), the second I find harder to justify.

> Would you rather trust a human legal system

The real world is far from that clear cut. Users of other private currencies, such as eGold, might be tempted to disagree. Hacks (in any and every sense) of those systems were even more devastating.

If you only ever consider every currency an alternative to the dollar, and limit yourself to that use case, you might as well ask yourself why any other currency exist at all.


I do think he contradicts that sentence in the following one:

> ...and criminals who want a black-market way to exchange money.

So not "useless" but of questionable value. But I believe bitcoin will probably never go away for this reason alone.


There will probably never be an economy or society that exists without a black market but that doesn't mean aiding such markets is a feature, not a bug.

Well it depends on the black market.

It might not be a feature in your opinion, but other people disagree, and do believe it to be a feature.

There are literally billions of people in the world that currently live in what I'd describe as an authoritarian dictatorship. So yes, I support black markets in these areas.


It's a feature. The black market economy is part of the overall economy, and at least a double digit percentage of it. Elimination of black markets means a massive overall economic loss for a country.

No government wants to completely eliminate all black markets. They just want to control them enough that they're taking their slice of the pie, and prevent certain markets which involve things like violent crimes which have bigger societal effects.

"Crime" without victims is wasting law enforcement resources. Tax money should be spent fighting real problems and not the virtual problems made up to have a perceived moral high ground.

Government over-regulation, or over-enforcement of regulation leads to economic stagnation and has the opposite intention of pushing people towards black-markets. It's a delicate balance.


One mans "criminal" is another's political activist living under a dictatorship.

I'd argue that a majority of the world's population currently live under what could be described as an authoritarian regime.


When that trust turns out to be misplaced, there is no recourse. If your bitcoin exchange gets hacked, you lose all of your money. If your bitcoin wallet gets hacked, you lose all of your money. If you forget your login credentials, you lose all of your money. If there's a bug in the code of your smart contract, you lose all of your money. If someone successfully hacks the blockchain security, you lose all of your money. In many ways, trusting technology is harder than trusting people. Would you rather trust a human legal system or the details of some computer code you don't have the expertise to audit?

In making the case for human systems over bitcoin, Bruce Schneider echoes the same arguments that eventually led to FDR stealing every US citizen's gold in 1933 to bail out the federal reserve. To me it seems deplorable for someone to advocate placating dishonest institutions founded on a legacy of screwing over people.


I don't read it as making a case for human systems over bitcoin. He is simply pointing out that bitcoin (and all cryptocurrencies) are more of a human system than the true believers believe.

The point of Bitcoin is that the trust is not in specific humans or groups of humans, but in a set of humans so diverse and distributed that there is no conceivable way for a minority of them to collude to screw over the other users. Moreover, they are deterred from attempting to screw over others, because they must exhaust massive amounts of electricity to attempt to cheat them, which can't be returned whether they succeed or not.

Unless the miners or mining groups collectively decide to fork.

I found it interesting that one of the first things that happened as bitcoin grew was the creation of mining collectives - literally the opposite of decentralization.


> Unless the miners or mining groups collectively decide to fork.

Miners aren't running the show by themselves. Monero will soon fork away and brick all ASICs again for example.

Forks aren't opposite of decentralization. In fact it's decentralization at work when different communities can split off and do their own thing. You're always free to use whichever fork you want, or all of them.


It's a mistake to look at pools as distinct entities in themselves or to think that their members are in cohesion to do anything other than increase their chance of monetary gain from their pool generating valid bitcoin blocks. The mining collective has no power of its own, but its power is distributed entirely among its members. A mining operator loses no autonomy by joining a pool, and can leave it just as easily as the joined. Nobody is loyal to a pool.

This is a supreme "Homo economicus" argument - there are switching costs, human factors (do I like X, do they pay on time, are they charismatic and forceful?), and dumb money problems that thwart it.

Not only 1933, they've been stealing money ever since through a monetary policy of constant inflation. Deliberate inflation is theft of a small percentage of the money of all savers.

You do realize that non-inflation was one of the main causes of the great depression, right? Are you telling me that you would spend your money on consumer products and employing people if you knew that it would be worth more if you left it sitting in your bank?

I'm not arguing for non-inflation, but I'm spending money on consumer products all the time instead of investing it.

What I believe is that if you want to use an inflationary currency, then that should be your choice.

And if you disagree, you should also be free to use a deflationary currency.

Both groups should be able to choose which currency that they want.


A currency cannot be simultaneously deflationary and functional as a means of exchange facilitating commerce. The laws of economics don't bow to your personal preference.

This is simply not true. If deflation is low, it is not a big issue, much like low inflation isn't a major issue. Hyperdeflation causes problems, as does hyperinflation.

The arguments against deflation don't really add up. There are numerous examples of deflation working in economies today. Example: Cars which lose 20% of their value upon purchase, and around 10% per year until they're worth scrap metal.

If the Keynesian economists are to be believed, then nobody would ever buy a car! If they just held onto the money, it would be worth more than the car will be worth 1 year later. What reason would anyone ever have to invest in a car? For some odd reason, nearly 80 million people do it every year.

Computers and Moore's law are another example. Why would anyone buy a computer now when they can just wait 18 months and buy one twice as powerful?

People change their spending habits when deflation occurs, moving from high time preference to a lower time preference, but investment doesn't stop. The most savvy investors will take the opportunity to try and increase their assets by productive means, so as to compound the effects of the deflation.

If people end up out of bullshit jobs, perhaps they could learn to code or something, instead of begging to state to steal other people's money.


Lots of people have used deflationary currencies all through history, actually.

Also, your point is irrelevant.

If you want to use an inflationary currency, you should totally be free to do so.

I merely think that there are lots of other people who would prefer a deflationary one, and would accept both the pros and cons, and that they should be allowed that choice.

May the best currency win! If inflationary currencies are so awesome, then you have nothing to worry about ;).


The U.S. dollar was deflationary, excepting the 1812 and Civil Wars, from the 1780s through the 1890s.

http://1.bp.blogspot.com/-QFYk6fgnd7c/UOr3hXo3LNI/AAAAAAAAkq...


Whether you think inflation is justified or not does not negate the argument. It is still theft. If savers did not have money, economists would not have anything to inflate. The inflation benefits the money printer, because they spend it at the market rate at the time of printing, but by the time it works its way through the economy, the effect is reduced purchasing power for the saver or consumer.

Yes, inflation is a deterrent on saving, pushing people to instead invest their money. Whether it is a good thing that people blow all of their money on junk consumer goods is another question.

What if some people just want to save, and not take risks? Well, it's their own choice, and now they are able to make such choice because there are no economists controlling the monetary policy of Bitcoin. What you think was justified previously, might no longer be possible. If we slip into depression due to deflation again, we might have to come up with some other solution, which isn't just kicking the can down the road.


I consider this debate an unresolved issue in economics. We don't know how to create a genuinely balanced and productive economy.

Option #1: deflationary economy where monetary velocity collapses and with it employment, investment, and innovation, ending in permanent depression (dark age) and slavery to a money-hoarding landed rentier class. Board game: Monopoly.

Option #2: inflationary economy that prioritizes spending and investment over saving, runs on debt, and ends with slavery to consumerism, bullshit jobs, and landfills overflowing with junk. Board game: Risk.

Pick your poison.

In the 20th century most nations went for option #2 because those that didn't tended to be overcome by the financial and military power of those that did. China is currently performing the most extreme inflationary economics experiment in history.

This post makes me remember a definition I heard once in school of economics: "the science of why everything sucks."


I mean it is resolved, and your two options kind of show it. The resolution is very low and slow inflationary economics, to match the growth rate in the economy. This should prioritize a safe level of saving and investment.

Higher and lower rates of inflation can be enforced (via central banks) to help fight recession or demand driven inflation (price level increases). However constantly tweaking inflation rates to get the exact right level for the current period of countries economic cycle (debt/savings levels, price levels etc...) is quite difficult. So yes we do know "how" to balance an economy in a general sense, in the same way we know how to build a web application in the general sense, but there is always work do be done in the details.


We've been trying to do that. The problem is that macroeconomic inflation is a big sausage number that doesn't really say much about the reality on the street.

Take a look at this:

http://thesoundingline.com/why-inflation-is-much-worse-than-...

Price stability has only been achieved for a few categories of things and the divergence is huge.


> What if some people just want to save, and not take risks? Well, it's their own choice, and now they are able to make such choice because there are no economists controlling the monetary policy of Bitcoin.

That's absurd. Putting money into bitcoin is about as risky as it gets besides vegas.


Inflation isn't really theft because:

- you knew it was part of the deal when you chose to hold fiat

- you can get interest on it to counter the inflation

- it benefits borrowers

- the money effectively goes to the government if not borrowers so it's more like taxation than theft.


This is technically a repost from the Wired article.

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


> Bitcoin might theoretically be based on distributed trust, but in practice, that's just not true.

I'll grant that Bitcoin is not perfectly centralized. But I challenge you to find a more decentralized digital currency on the planet.

A digital currency doesn't need to be 100% decentralized to be very useful as a digital medium of exchange.


It depends on your definition of "useful". It doesn't need to be 100% decentralised to allow exchange to take place without a single technical point of failure existing, but it probably does need to be substantially more decentralised than it actually is to not actually have bigger trust issues than traditional currency (at least for people trying to carry out legal transactions in a developed world financial system).

> In fact, a system where you can lose your life savings if you forget your key or download a piece of malware is not particularly trustworthy. No amount of explaining how SHA-256 works to prevent double-spending will fix that.

A second-layer solution, built on top of bitcoin, that would allow transactions to be reversed by a central authority, if a person decided to place their trust in that authority, could fix that, though.

This is the problem with the article: it doesn't take into account bitcoin's ability to change. It's like Schneier is saying the internet will never work because dial-up modems are slow and hog your phone line.

He's assuming no further innovation will occur. And like the early internet, it's practically impossible to predict what cryptocurrencies will look like in a few decades.


Of course trust is still involved. The whitepaper specifically puts "trust" in the context of trusted third parties ("financial institutions serving as trusted third parties to process electronic payments"), and proposes a solution to the double-spend problem when they are not relied upon. It is a very unique solution that could be useful in solving some very specific use cases. There are a wide number of risks that could cause bitcoin to crumble - Core Dev Matt Corralo often says he gives bitcoin a 5% chance of success. The wide-sweeping conclusion is that blockchain is useless, though I would agree in almost all cases, it disregards the successful instances where trust minimization provides substantial benefit.

>but centralized (as in there's only one)

I think that should be "global", especially since "centralized" means something significantly different in blockchain circles (and bs goes on to use it in that sense later in the article).


Logically centralized but physically decentralized would be a specification, but his statement is correct. "Centralized" does not have a different meaning in blockchain circles, however it (and especially the inverse "decentralized) has become a weasel word that people use frivolously.

I've definitely seen, and been guilty of, using "centralization" as a weasel word in blockchain circles. Agreed this should change. Usually what's meant is "potential for corruption or abuse."

"Centralised" is usually used in the blockchain space with the systems theory definition as applied to graphs. Ditto for "decentralised" and "distributed"

If blockchain is so great, where are all the killer apps for this fantastic new technology? C'mon, we're 10 years into the cryptocurrency revolution, and we're still arguing about hypothetical use cases.

I think ICOs were something of a killer app and they served to move things forward considerably. I don't think people have fully realized the breadth and depth of impact ICOs have had yet.

For some relativity:

  - 1969: ARPANET
  - 1971: Email
  - 1973: TCP/IP
  - 1980: Usenet
  - 1989: Commercial ISP
  - 1994: Netscape 
  - 1999: AJAX / Live Journal
  - 2001: Dot com bubble.
  - 2004: Gmail/Facebook
  - 2006: Twitter
  - 2009: Bitcoin
  - 2011: Twitch
  - 2013: First ICO
  - 2014: Alexa
  - 2015: Ethereum
  - 2017: Cryptokitties, EOS
  - 2018: Tron
  - 2019: Profit?
I think the next wave of killer apps will include a lot of creative new games. I think gaming will be a big force in blockchain adoption in general.

Nice timeline.

But... Tron?!


I'm not a Tron fan, I listed it because 1) it was something that happened last year and 2) a few of the most played dapp games are running on Tron right now (and EOS)

Now I'm wondering, what else happened last year? I was sleepy.


"Blockchain" is over-hyped. Bitcoin is the real innovation. The use-case has been clear since its inception: It is censorship resistant digital money with no externally-decided monetary policy.

There are a lot of people who want it to be something else, or who want to sell you some snake oil because they missed out on the opportunity to accumulate a lot of Bitcoin early on when it was cheap, and think that they'll be able to repeat the process with them at the forefront. It is all utter nonsense. Each project which prints its own token undermines whatever technology they think they've invented, because the fixed monetary supply was one of the key innovations of Bitcoin. Removing that innovation from your own project only makes your project inferior to Bitcoin by design, but bitcoin still remains immutable to changes in its monetary policy.

Every "blockchain" idea can actually be implemented as a pegged sidechain to Bitcoin, and avoid the need to print its own currency. This is the simple litmus test of whether or not it is snake oil.


To me, Bitcoin seems like an ecological crime that abets other crimes.

To me, the internal combustion engine seems like an ecological crime that abets other crimes.

Yes, ICEs also needs to be eliminated as soon as possible. Switching off bitcoin is likely the easier of those two transitions.

If you're advocating that we innovate an alternative to bitcoin, which has the same decentralisation and censorship-resistance properties, but a greatly reduced energy footprint (for the same level of security), then I agree and support that transition.

If, instead, you're suggesting that some (global?) central planning committing should decide which technologies are allowed to exist, based on some subjective determination that a given technology uses "too much" energy compared to how "useful" it is, then I think such a system of control is likely to be abused.

For comparison, 2nd generation cryptocurrencies (like Litecoin and Ethereum) are comparable in their energy usage (per unit of economic value created) to mining rare earth metals, and over 10 times more energy efficient than producing aluminium:

https://s3.cointelegraph.com/storage/uploads/view/728accd677...


No better alternative is going to be found for Bitcoin. Bitcoin's security depends on it being super expensive to attempt to hack. Electricity is perhaps the most in-demand good (aside from food and water) over the globe. Those other systems like PoS do not have the same irreversible commitments as the commitment to exhaust energy. No matter what you do, you aren't going to get the energy back.

Bitcoin miners will force innovation in cheap renewable energy sources because without them, they will be out-competed and will not stay in business for long, because someone else will make the investments. Bitcoin mining will eventually be run on low cost, low maintenance renewable sources. Some of it already is. Real business needs end up producing results. Virtue signalling middle classes who won't put their own money where their mouth is do not.


The Chinese government could effectively end bitcoin mining tomorrow if they decided to.

Aluminium is useful. Rare earth metals are useful. Bitcoin (see top question) is not useful (citation: after 10 years, almost nobody is using it).


> The Chinese government could effectively end bitcoin mining tomorrow if they decided to.

So you do want a central planning committee to control which technologies are allowed to exist in the world. ;-)

You're right, though, the Chinese government could greatly reduce the global hashing rate of bitcoin, which would make blocks much harder to mine (temporarily), and slow transactions down. The network would continue operating though, and perhaps Iceland would become the predominant location for bitcoin miners:

https://www.theguardian.com/world/2018/feb/13/how-iceland-be...

With their clean energy supplies and greater internet freedom, that might not be such a bad outcome (except for China's economy, potentially).

> Bitcoin (see top question) is not useful (citation: after 10 years, almost nobody is using it).

The top comment I see in this discussion is:

https://news.ycombinator.com/item?id=19146037

and it contains multiple examples of use cases. Maybe "almost nobody" cares about WikiLeaks, or weed, or porn, or Venezuelans, or privacy, or having an alternative to PayPal, but "almost nobody" needs a cure for a rare illness, that doesn't make it useless. Similarly, nobody actually needs Facebook (or Hacker News, for that matter), which also uses a lot of electricity, so if we're going to start banning technologies for being not useful and using energy, then why don't we start there?


Are you kidding me, or just kidding yourself here?

Orders of magnitude more people care about weed and porn than care about bitcoin. Your comparisons are all ridiculous. Bitcoin is like curing a rare illness? This is why nobody takes breathless crypto-boosters seriously.

Bitcoin is MLM for tech boys.


You seem to be arguing against positions I'm not taking, so perhaps I didn't make myself clear.

I don't dispute that orders of magnitude more people care about weed and porn than care about bitcoin, but the reason I mention weed and porn is the same reason "lawn" did in the comment I linked: There are people who wish to buy or sell these legal (in some jurisdictions) products but who are limited in their practical ability to do so due to a lack of financial freedom. For such people, crypto-currencies are a possible solution.

https://news.bitcoin.com/indospace-io-cannabis-merchants-acc...

https://motherboard.vice.com/en_us/article/vbqwwj/patreon-su...

https://bitcoinist.com/librepatron-btcpay-patreon-bitcoin/

The fact that bitcoin is used in these situations (even if for a minority of purchases) shows that it is a useful technology (assuming you think that these products aren't inherently illegitimate, even where legal).

Similarly, I accept that bitcoin is completely different from curing a rare illness, in terms of the effect it has on its beneficiaries, but that was not the comparison I was trying to make. I was trying to point out that just because only a minority of the population benefit from a technology (whether a medical technology, or a financial technology), that doesn't make the technology useless.

To pick, perhaps, a better example, suppose there were a new financial technology that helped only the poorest people (and I'm not claiming that bitcoin is such a technology). A wealthy critic of the technology might claim "almost nobody is using it", and they might be correct, but they would be missing the significance of the technology.

Full disclosure: I have never bought any crypto-currency, and have no intention of buying any in the next 24 hours. :-)


Scammy investments have been huge! Many of the remaining enthusiasts think this will transition to non scammy investment via security tokens.

The trust Nakamoto refers to is explicitly and obviously the trust demanded by "trusted third parties" that intrude into transactions.

Removing this intrusion allows increased trust between members of society.


This person is unfamiliar with what it means to run a public trading vehicle along with the value that comes with that. Public trading vehicles are designed to exercise capital from the capital markets. This is what enables public companies to grow. Crypto another form of a public trading vehicle in a new global capital marketplace and it helps legitimate companies grow.

When that trust turns out to be misplaced, there is no recourse. If your bitcoin exchange gets hacked, you lose all of your money. If your bitcoin wallet gets hacked, you lose all of your money. If you forget your login credentials, you lose all of your money. If there's a bug in the code of your smart contract, you lose all of your money. If someone successfully hacks the blockchain security, you lose all of your money. In many ways, trusting technology is harder than trusting people. Would you rather trust a human legal system or the details of some computer code you don't have the expertise to audit?

Most of these things have nothing to do with "blockchain" or Bitcoin. Exchanges are outside the scope - entirely. They are merely one kind of merchant. "Wallets" are also outside the scope, because the protocol only knows about scripts. Login credentials play no role whatsoever in Bitcoin - that's an add-on provided by third parties.

Bugs in contract code are a problem. That's why you don't trust it. You verify it - just like you read any contract you sign.

"Hacking blockchain security" is another matter. What Schneier is referring to is the Ethereum Classic majority hash rate attack. That's a well-known limitation of all block chains. Satoshi's white paper deals with it extensively and quantitatively. It's not like Schneier has stumbled onto something novel here.

These four elements work together to enable trust. Take banking, for example. Financial institutions, merchants, and individuals are all concerned with their reputations, which prevents theft and fraud. The laws and regulations surrounding every aspect of banking keep everyone in line, including backstops that limit risks in the case of fraud. And there are lots of security systems in place, from anti-counterfeiting technologies to internet-security technologies.

The nativity here beggars belief. Does this man not remember 2008? Does he not remember HSBC? Does he not know that the individuals responsible for the most egregious bank fraud have never been punished in a criminal sense? Does he not remember the vast amounts of sensitive, personal information leaked by these institutions? Has he never heard of the massive, unconstitutional civil asset forfeiture problem in the US?

The best you can say about the banking system is that laws/regulations may inhibit, to some degree, some forms of theft and fraud.


The loses as a percentage of funds have been vastly less in conventional banking and similar than in cryptocurrency.

Regressive author can't understand the Theory of Conservation of Mass. Gives us the environmental waste straw man and follows it up with an appeal to power. Fails to understand that the real innovation in trust is the stable horizontal structure that makes top down governance obsolete. Enjoyed reading anyway

> is the stable horizontal structure that makes top down governance obsolete

lol, how can bitcoiners still believe this? SegWit was introduced top-down, The DAO was top-down governance.

Everthing about bitcoin, the mining, the software, the majority owners of all bitcoins ever mined, is ruled and in the hands of less than 12 people.


The counterpoint that those n<12 people are supported by, and theoretically accountable to, the vast ecosystem of miners, investors, and developers.

...and the counter-counterpoint: for all the shortcomings of modern democracy, the same is at least somewhat true of the board members of the Federal Reserve, and the elected congress-critters who occupy the relevant regulatory committees.


what do you mean by ruled? there is no ruler. thats the point

Spot on analysis of the logical fallacies this author relies on to make their point. It's too bad, because his analysis and summary of many of the features of blockchain and cryptocurrencies is quite good.

This analysis makes for a great checklist for auditing projects to find the rare handful that have identified these same problems and are working to resolve these issues. I don't understand why critics with this level of understanding insist on binary assessments of performance or use the "10 years is a long time" argument. The bubble bursting has been effective at clearing out the scams and garbage. There's a long way to go, but bathwater is easier to drain these days too.


almost all the comments are negative about the article but many people upvoted. I like the dynamics of HN

Increasingly I have noticed that people who can see the future have given up on engaging those that can not.

I like Bitcoin, I don't see a use for any other coin. Decentralized verification seems to be valuable in Currency only.

Timestamps and voting do not seem to be valuable, this was the craze over alt coins.


This article uses the common criticism of there already being existing institutions and security systems for money transfer. However, this may be true in well developed countries but it's not true at all in developing countries. IMO, this is a common oversight by critics.

In what world do people have the technological and social infrastructure for Bitcoin (end user devices, networks, miners, etc) but bad infrastructure for money transfer? How do you think they would be buying their phones/devices and paying to connect to the network?

You act like only people in rich and peaceful countries can have phones and access to internet. All you need is a smart phone or a laptop with a very poor internet connection to use it.

Everyone doesn't have to be a miner. Yet there are many miners in Venezuela, a country where Bitcoin can have a very large impact for people.

Venezuela have very bad infrastructure for money transfer... Since it's all blocked. Except for Bitcoin.


Venezuela for one.

How? There's always a flavour of the moment failing state inserted as a reference here. But if you can't transfer Venezuelaean currency out of the country, how are you buying bitcoins with it? How are you getting the money to that person?

Check out LocalBitcoins

Don't worry. If there is a developing country where those institutions don't exist, then you can trust the CIA will forcibly install such institutions, and every public mouthpiece from Wall Street to Capitol Hill will tell you why that's a Good Thing.

> It can also be extremely expensive, both in data storage and in the energy required to maintain it. Bitcoin has the most expensive consensus algorithm the world has ever seen, by far.

When Bruce complains that Bitcoin's consensus is expensive he really compliments Bitcoin's security and resistance to attacks on immutability.

He's spot on that private blockchains and most public blockchains are nothing but hype, that opinion is not news though.




Applications are open for YC Summer 2019

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

Search: