Ask HN: What was impossible earlier that is easy in the BTC and ETH blockchains?

60 points by noloblo ↗ HN
What was impossible to build a few years ago has become incredibly easy if not simple in bitcoin/ethereum blockchain?

If you were to rebuild any of the consumer facing web apps/web infrastructure technologies which would you rebuild on top of the ethereum blockchain

71 comments

[ 3.1 ms ] story [ 137 ms ] thread
By no means impossible before, but now much easier: Dark net markets, money laundering, unregulated securities offerings, ponzi schemes, scams, ransomware, and, to a lesser extent, gambling.

Since you asked about rebuilding web apps, I'll link you to patio11's comment, which leads with "Blockchain is the world's worst database"

https://news.ycombinator.com/item?id=13420777

Spot on, these along with: political donations, funding activism, information marketplaces (think wikileaks but the leaker gets paid).

Anything that a government or regulatory body would want to censor, basically, is made significantly easier with Bitcoin/the blockchain.

patio11 was entirely wrong about bitcoin.
I don't see that anything has changed since his comment
It wasn't impossible before, but with BTC/ETH it's far easier to launder money. With a Cryptocurrency tumbler [1] it's almost untraceable.

Previously, you had to carry the cash yourself, or use a service to transfer the wealth for you (through falsified business transfers, casinos, diplomats, flight attendants, etc.) Now you don't need all that, you can practically do it yourself. There are less people involved and it's harder to trace.

[1] https://en.wikipedia.org/wiki/Cryptocurrency_tumbler

no, this is just plain wrong. the amount of controls in place, banking hosyility, and the market liquidity make digital currency terrible options for laundering.
When compared to previous methods, laundering through BTC/ETH is far easier. With a tumbler, or better, several tumblers, it's nearly untraceable too.
until you cash in or out... which is a key part of money laundering...
Compared with cash, digital currencies are much harder to launder. It will be obvious when you try to cash out that the money went through a tumbler. With cash, you always have plausible deniability unless you were physically seen receiving it in some illegal manner.
ICO.

ICO is Ethereum's killer app - and that's ok for now.

Excellent question, I'd really like to know as well, to see what's beneath the hype.

What I do know: anonymous money transactions. Of course, you'll still have the anonymity loss when converting to/from dollars. While the parties are anonymous, all transactions are visible. In many cases that means that people can end up associating your bitcoin address with your identity (assuming they manage to puzzle out the patterns of your transactions and cross-check that with known addresses). Mechanisms like CoinJoin allow you to mix your transactions with those of others, in order to obscure them, but then the CoinJoin provider still sees your transactions.

Look into zkSNARKs which is what for example Zcash uses for completely anonymous transactions without any mixers or such.
Escrow with shared programmable ownership, global value transfers completed in less than a hour, immutable public timestamps, self-owned online identity etc All impossible pre-blockchain
> self-owned online identity

gnu-social existed long before blockchains, as did many other federated systems.

> shared programmable ownership

What does that mean?

> global value transfers completed in less than a hour

Customers of the same bank can do the same globally usually. It happens they're transferring "internal bank numbers" instead of real money, but all cryptocurrencies are also just proxies for money and both sender/receiver must be using the same crypto-currency. Not very different.

Also, paypal, snapcash, etc seem to work just fine without a blockchain.

> immutable public timestamps

The library of congress begs to differ. So does gnusocial (federated/distributed timestamp). Mailing lists were used for this purpose at some point too.

Things like freenet exist just fine without blockchains.

This is a good list of inputs, though I think immutable public timestamps is the only one that was impossible pre-blockchain, and it's not clear what the value of this is.

I'd argue that the rest of them were not impossible before blockchain. From my perspective blockchain just makes these easier/cheaper.

This is not a judgment of BTC/ethereum, but any service which wants to transmit value and protect against repudiation but avoid government scrutiny is vastly easier with BTC/ethereum. This includes Silk Road and most modern ransomware.
Despite the blockchain hype, "money" is the killer app for bitcoin and ethereum.

Before bitcoin, an online business would be taking a huge risk if it accepted a credit card payment from someone in Latvia. The fraud rate on such payments is probably 50%. With bitcoin, it's safe and fast. I own such a business (a VPS hosting company) and Bitcoin really has been a great solution for us.

Before bitcoin, a legal pot shop would have to deal with cash because no bank would give it an account. Having all this cash around creates a danger for the employees, not to mention the armored car drivers. Bitcoin is a tool that can be used to solve the problem.

I definitely wouldn't say "fast". I can tap my phone with apple pay or insert my VISA chip card and it takes <5 seconds.

I agree though that for payments where the merchant can't be banked or VISA doesn't want to support them, Bitcoin or cash are the only other options.

the <5 seconds is not settlement. It's a 3rd party signalling to another 3rd party.

Bitcoin confirmations are just that, confirmations of transaction being approved by the network.

-blockchain is immutable -only gets written if valid -takes 10~60min to be confident of the reality -no chargebacks

As a user, <5 seconds is much faster than 10~60 minutes. And I have never had VISA mutate my transaction... that would be fraud.

Once every 200 transactions let's say, my card will be denied and I'll need to swipe it a second time. This is significantly less of an ordeal than having to rely on the inefficiency of the blockchain every single transaction.

It does not take 10~60 minutes to sign a bitcoin transaction... It doesn't even take a second. Maybe a slow PoS will need another second to actually verify the transaction, but this is still way faster than VISA.

I regularly spend more time than that on VISA transactions because I don't live at my billing address on another continent.

> It does not take 10~60 minutes to sign a bitcoin transaction... It doesn't even take a second.

Why are you only considering signing time? Signing a bitcoin transaction is not sufficient for the merchant to be confident that they can give you your goods. You must submit it to the network and the merchant must be reasonably confident that it will be accepted in a block as-is and that the block will not be orphaned.

Edited the comment to include verification almost simultaneously with your post :)

But yeah, obviously as the transaction value goes up the merchant will need to wait a few more seconds to counter doublespend attacks.

>You must submit it to the network

Not necessarily true, ideally the customer doesn't need internet connection and just transmits the signed transaction to the PoS.

>and the merchant must be reasonably confident that it will be accepted in a block as-is and that the block will not be orphaned

First of which can be verified very fast by a computer. I believe orphaned blocks are rare enough that they're mostly a non-issue.

I guess it's important to point out that in practice most Bitcoin ATMs I've used did 0-confirmation transactions up to 1000 euros after only a few second wait, so it's not like all bitcoin merchants suck.

The average confirmation time for a Bitcoin transaction was 400+ minutes recently (and is still around an hour even now). https://blockchain.info/charts/avg-confirmation-time?daysAve...
Yes, but you don't need more than 20-30 seconds to be confident that a transaction isn't a doublespend attempt.

After that it's only about checking if the fee is high enough.

> After that it's only about checking if the fee is high enough.

Which can take up to 72 hours, right?

No, you just check if the fee per KB is more than x. Where x is a fee that is likely enough to confirm. This is practically instant.
I'm not sure "likely enough to confirm" is going to go over all that well with many merchants, particularly when that chart shows the price-per-byte jumps all over the place rapidly.
physical mail through the post office still exists alongside e-mail.

Just like credit cards will exist alongside cryptocurrency. There already exists merchant processors like BitPay that allow for 0-conf transactions. I'm not saying that's ideal or the permanent solution, but because bitcoin/crypto is software, it will continue to be improved upon over time.

Is your argument that the VISA network is not software that is being improved over time?

Did you know that VISA has been able to process up to 56,000 transactions per second in 2014? It also sounds like they improved 20% from 2013 to 2014 [1]

If the past 8 years are of any indication, bitcoin is more like physical mail than VISA is in your analogy, in terms of performance, ease of use, and how fast it is improving.

[1] http://www.digitaltransactions.net/news/story/Visa_s-Test-Re...

One market is international peer-to-peer payments.

My US credit cards work well in most of the world if I want to buy from some kinds of business (ex. use a chip card to pay for gas, hotel, groceries, restaurant meals, etc. in Canada.)

If I want to send money to a bank account elsewhere in the world my credit union charges about $40, and most other banks charge about the same. This is not so bad if you need to move $100,000 but the fee is too much to move $100.

Bitcoin is not as fast as swiping a credit time but bitcoin and a wire transfer are on the same scale, and even if you pay a $5 miner's fee on a transaction that beats the wire transfer.

Remittances have been shown to not actually be better with bitcoin. [1] [2] [3]. I can send 2000 USD to India using TransferWise and they will only charge me 17.84 USD. That is a 0.89% fee using a reliable mechanism and I know my recipient will have their native currency in their bank account within a couple business days. Compare that with opening an account on a US bitcoin exchange, sending money there (time delay), buying bitcoin (exchange fee), opening an account in india (requires verification and linking a bank account), sending the bitcoin over (time delay), converting the bitcoin to rupees (time delay, exchange fee, volatility risk), then sending the money to the bank (another time delay).

[1] https://medium.com/cryptonight/bitcoin-doesn-t-make-remittan...

[2] https://bitcoinmagazine.com/articles/rebittance-startups-agr...

[3] https://www.saveonsend.com/blog/bitcoin-money-transfer/

Roll back.

A Bitcoin transation is not a credit facility. This is an important differentiation.

When you pay with a credit/debit card, the money isn't transferred right away.

Visa/MC check to see if they are willing to approve the transaction(see if you've breeched your limit)

if they are, they guarentee that you are good for the money, and the merchant approves the transaction.

A few days later, visa will settle with the merchant, paying your debt on your behalf.

You then pay visa on your monthly schedual.

Visa also act as an escrow of sorts, If there is fraud, or the transaction becomes void, Visa is equally liable to sort it (depending on country).

Bitcoin has none of these functions, it is just digital cash, with a public ledger. Whats worse is you need a credit function to allow quick purchasing of thing with bitcoin, because of the time needed for verification.

Without a relable and safe credit/escrow system digital currencies will just be limited to the brave or those who have a decent cradit facility.

> This is an important differentiation.

From the user's perspective, not really. Many people compare their credit card to Bitcoin even if the way these systems work are completely different (credit network atop fiat versus digital asset, as you say).

At the end of the day, all that matters is how fast, secure and reliable these things are to the user. If they can pay for their loaf of bread within 5 seconds and easily pay off their credit card once a month, and the Bitcoin alternative is using a crypto-exchange, exposing yourself to volatility risk, all for something that is slower and less reliable at the point-of-sale... Bitcoin is far behind.

Sure, if we had a credit system on top of Bitcoin, it would probably have an easier time competing with VISA credit cards, but the fact is that does not exist so cannot easily be compared today.

Maybe to be more precise "anonymous cash replacement" is better?

Electronic "money" in general has long been available through credit cards, PayPal, etc but at higher cost and loss of anonymity

Right, but it's more than electric money. The OP mentions it's electric money that:

1) has assured/immutable transactions (unlike CCs) - This may or may not be a good thing from a purchaser perspective on seller fraud.

2) is available to all individuals (unlike banks) - Again, which may or may not be a good thing if you're worried about money from/to the grey/black market.

You've just shifted the risks from the business to the customer. As a customer, my credit card gives me significant protection from paying for a VPS from someone who intends to scam me.

Somehow, I suspect more people have had Bitcoin keys stolen by malware than have had their armored cars robbed.

"Money" is not a single app, it is three apps in one. It's used to store, transmit, and account for value.
And I would argue that cryptocurrencies do none of those things. What value is being stored? Owning some BTC just implies that someone at some point burnt some electricity and computing power to solve a rather insignificant crypto problem.

This is precisely what I don't get about cryptocurrencies. There is no intrinsic value, and they're not even backed by the full faith and credit of, well, anyone. I don't see why it's "worth" anything.

And, we don't need a digital currency. We already have a bunch of them (the US dollar among them). Most money in the world doesn't exist as physical coins, notes, gold/silver, etc. It's all bits in a computer, but those bits are backed by a sovereign entity. That's why we are able to use them as currency.

If anyone can explain the fundamentals of why BTC or any other cryptocurrency should have any value whatsoever, I would really appreciate that.

I think their intrinsic value is backed by their usefulness as a medium of exchange for the use cases described in the parent comment. Solving the crypto problem seems more like something that makes them useful for exchanging.
Solving the crypto problem is just part of the mechanism that ensures there's a limited supply of them. We could do the same thing by randomly assigning floating point numbers between 0 and 2.1x10^10 - 1 (corresponding to the 21M BTC that are ever allowed to exist).

The part that makes BTC relatively safe to transact with is the blockchain itself.

Exchange bitcoin with real money and ask yourself the same questions.

> And, we don't need a digital currency. We already have a bunch of them [...]

That's a bit enraging to me. I feel it's in the same class of arguments with things like earth is flat -- we don't want a round one, blacks are not humans -- they should be happy we grant them slavery. The status quo is our normality.

> backed by a sovereign entity

Which one? I had savings to buy several apartments but in a few years those savings were worth a lunch. That was a long time ago. But as of now my income depends on US, EU, Syria, Turkey etc crisis and hell knows what other stuff I don't even hear about on the news.

As a guy lacking both bitcoins and real money they both weight as a surprisingly similar amount of crap.

>It's all bits in a computer, but those bits are backed by a sovereign entity. That's why we are able to use them as currency.

I've always felt this argument is disingenuous. People traded goods and services prior to government-backed currency. Being backed by a government sure does simplify things and make more people value it but it isn't necessary. Government-backed money can turn into Monopoly money as well.

Lest you forget 1920's Germany [0].

[0] http://i.imgur.com/nk0CnNr.png

tokenization of assets of any variety. some people will say ICO but its bigger than that.

People tend to miss the bigger picture when it comes to digital currencies. Bitcoin and the blockchain are communication protocols. They allow anyone and everyone to communicate and transact in a single unified language, making it easy for people to put something out into the world. Wether its a purchase or a remittance or a smart contract executed ICO. the currency themselves are just tools for make global statements.

Bitcoin demonstrated that is is possible to dis-intermediate the storage and transmission of value, an area that is the mainstay of the banking industry.

Ethereum allows for open experimentation, and has brought about the emergence of arbitrary tokens.

Decred has demonstrated that is it possible to dis-intermediate the process of political decision-making for a cryptocurrency.

Does Bitcoin's simple use case of a scarce digital commodity count? It's not that useful in the US/Western Europe but is genuinely useful in certain countries like Venezuela where inflation is in the triple-digit percent a year.
A scarce commodity isn't new though right? People can buy gold online now and use that as a safe haven, for example. Why does it being a "digital" scarce commodity matter?
Is there actually a healthy bitcoin economy in the meatspace of Venezuela. Can you trade it for things? Can you buy milk with it? If not, it's really not much of a currency.

If you are just looking to hedge against inflation, you can convert to any other currency or asset just as easily, if not more easily. I assume you can buy gold ETFs in Venezuela.

You will certainly get much farther there with Brazilian Reals or Peruvian Pesos.
Decentralized naming -- Bitcoin solves Zooko's triangle.

https://en.wikipedia.org/wiki/Zooko%27s_triangle

Hrr, I see the Wikipedia page says that Bitcoin doesn't provide "human-readable". But I think it does. You can register names on the blockchain and assign them to bitcoin addresses using OP_RETURN comments, with every party agreeing on who knows which human-readable name, despite having no centralization.

I'm not sure what you mean by "Solves Zooko's Triangle", as the Triangle is an assertion of a problem (trilemma) that can't be solved. If you mean "refutes", I disagree. Besides the problem with names, bitcoin is, of course, vulnerable on the security side---to 50% or more of computing power being used maliciously (or being fraudulently converted to hidden malicious use). While the distributed nature of bitcoin strongly mitigates this once the network is big enough, the point of Zooko's Triangle is that you have to accept no better than mitigation of risk on at least one leg, not its elimination. The possibility of Sybil attacks also point out issues associated with the anonymity of other network participants and the possibility of hidden collusion.
If I had the resources, I would rebuild all of the consumer-facing web apps using blockchain-based smart contracts. I expect that all web apps will be rebuilt in this way to an extent, either from within or by way of competition.

So "impossible" is a charged word, as is "easy". I would say that it was previously far more costly and difficult to implement the following:

- programmable, highly-reliable, do-it-yourself financial instruments

- services that cannot feasibly be shut down

- verifiably revocable authentication

- data that cannot be modified or deleted

- functionality that cannot be modified or deleted

- complex business interactions involving numerous parties that can be verifiably traced, from start to finish

Disclaimer, because everyone loves to bring up The DAO incident: the above capabilities hold true unless the community decides otherwise after a ridiculously onerous debate. Even then, the unaltered data/code can still live on.

Investing in crypto-currency is investing in blackhats, password lists, botnets, spam servers, click farms and M(alware)aaS. in currency collapse of struggling small countries and tax evasion in larger ones. it's investing in money laundering, corruption, opinion manipulation, corporate and political espionage and election tampering. in ransomware and plain-old ransom. in child pornography and drug, weapon and organ sales. investing in CC is investing in the ugliest side of humanity, and business.is.booming. investing in human misery used to be elite exclusive, but now even you can do it!
Right. This all started with the first bitcoin. We had a paradise before.
What a load of FUD.
https://www.openbazaar.org/

An online store with:

No Transaction Fees

No Monthly Fees

No Listing Fees

No Bank / CC Required

Live Chat with Customers

Store Customization

Peer to Peer (no middleman)

.. no protection against dishonest sellers

.. no protection against dishonest buyers

.. no protection against dishonest website operators

I think it is less about what you can do and more about what it prevents. Blockchains are really an example of triple entry accounting. A transaction is local to you, local to the party you deal with, and public and (more or less) immutable to the public. If we used, for instance, a government-run blockchain to record all business transactions making "cooking the books" much less possible (no Enron).

So it is useful for no fraud payments (guaranteed payment) and no trickery transaction/accounting, which is literally all blockchains are when you get down to it. The downside is that the transactions are all public knowledge to anyone contributing transactions. But it could immediately be used to track government spending (yeah right) or non-profit spending which should both be more or less open books.

Using a distributed ledger like R3's Corda (https://www.corda.net) will do what you say: prevent "cooking the books" and make transactions public and audit-able.

What the blockchain does that Corda does not is: the ability for anyone to be a transaction validator. What this ability gives the network is censorship resistance. Basically because anyone can validate transactions, no central body can say what should and should not go in the ledger.

The problem it solves is offering both anonymity and security.
It is a very common mistake to look at btc and other cryptocurrencies through the lense of what you know money is right now, in the context of the infrastructure that exists around it. The real possibilities lie in the infrastructures that can emerge around this new technologies. Consider the fact that crypto is inherently agnostic about who or what holds it. You can be a person, an institution, a script, or an object (think self-driving, self-owning, self repairing cars), or think trafficlights you can pay to turn green for you :).

then there is the possibility of true microtransactions. What are the fundamental parameters of internet commerce in our day and age. I would argue that there is roughly three kinds of dominant transaction. One of purchases of goods/services (Amazon etc), software as a service (Spotify), and then there is the vast space of businesses that basically live of the advertising they sell through brokers such as facebook and google. There is really no alternative for this lower limit of transactions. With bitcoin you can imagine a world where people offer content online (say music) and offer consumers the choice of a microtransaction/listen or to consume the same content with ads, without having to give up control over their content to third parties such as spotify. The possibilities are really quite endless.

Today's bitcoin confirmation fee is 88140 satoshis (~$2) according to https://bitcoinfees.21.co/

Does not sound like a good idea for microtransactions.

Segwit and the Lightning Network will solve this.
Previously it would've been very difficult to convince complete strangers to hand over millions of dollars in a matter of minutes.
>"When inflation is at the target rate of 3% that means that the central bank will default on 3% of the amount of debt it guarantees within a year."

Thats patently false. Government debt is sold in the form of bonds. They make up the bed rock of the whole economy. Precisely because they can "print money" they don't default on debt. Government bonds are considered "zero risk"

Yes, a central bank can default on foreign debt, or if they can't print their own money (see greek banking problem.)

however this underscored the fundemental misunderstanding of how money works.

>"By offering short term credit to each other market participants will be able to engage in economic exchange without having to pay the cost of money.

>Because market participants will effectively be creating new money with every transaction they will be expanding the money supply and increasing economic growth with every transaction."

Thats litrally how the entire banking system works now. Thats how credit cards work too. Thats how B2B credit works. What is not tackled here is risk. Sure free credit is great, but how do you know if someone can settle the debt? In this (unrealistic) world, the credit ratings agency becomes the dominant force.

Speaking of which, there is no reason why a small buisness can't offer its own line of credit, without using a bank. However the reason why credit cost money is because there is risk. Interest is a function of risk and supply. Sure a coffee shop _could_ settle bills at the end of the week, but the cost of administering that system is non trivial. The risk of loss is also non trivial.

Bitcoin or Etherium does nothing to stop that.

In short, this article is flawed.

post script:

> "As decentralized digital exchange replaces central banks the effective money supply will increase dramatically."

this is also known as hyper inflation.

People are using bitcoin more than never before to receive their payments for actual WORK.

If you don't believe me, just ask Bitwage.