Hacker News new | ask | show | jobs
by markkat 2019 days ago
It makes me sad, because I expected a comment like this to be the highest on HN.

In the case of Ethereum, digital scarcity secured by a blockchain enables a turing complete state machine that the world can use. In the most basic terms, this will remove clearing houses for transactions of assets. In the long term this will lead to novel types of assets, and make ownership extremely liquid. Imagine using your phone to buy shares in a recording artist you just discovered, and selling those shares when they win a grammy. Imagine building a stream of passive income based on the shares you've earned in projects you've worked on throughout your life.

As an example, already, anyone in the world can buy property in the US: https://realt.co/

What the web did for infomation, Ethereum will do for value.

9 comments

For me at least, "Imagine the possibilities!" is not an answer to the question, "What's it good for?" Indeed, my experience is that it's almost an anti-answer.

For example, in the mid-aughts I was talking with the CEO of a well-funded "semantic web" company about joining. I liked the people and the tech was cool, but I just couldn't figure out how it would turn a profit. His answer was similar to yours, a "what couldn't they do?" answer. Turns out they never found an answer, and got acquired for their technology without ever turning into a real business.

As somebody who used to write software for financial traders, it's not clear to me that removing clearing houses is a step forward. They provide important services. And I'm even more skeptical that allowing randos to trade in opaque, unregulated assets is a good idea. Music industry actors, for example, have decades of experience in creative accounting and screwing people over. Real estate has a strong caveat emptor tradition, and property management companies are not exactly a watchword for fair dealing. So from you description, I'm mainly imagining people all over the world getting fleeced with no practical legal recourse.

As best I can tell, the cryptocurrency space seems hell-bent on rediscovering why strong markets and government oversight exist in the first place. The only demonstrated market advantage Bitcoin has in a decade of operation is in light financial crime like money laundering, capital control evasion, ponzi schemes, ransom payments, and outright theft.

I do hope these technologies turn out to be useful for something. But the "imagine the possibilities" routine worked much better when there was less of a track record. Now I'd rather examine the actualities.

You touched on the core problem with cryptocurrency, and it is not a technology problem.

The problem is that cryptocurrency and the community around it is absolutely crawling with scams to the point that it's hard to find things that are not scams. Bad drives out good. Even if the tech works perfectly and can be made to scale, if the ecosystem is full of scams it will never be used for anything but scams and black/grey market applications. Anyone not looking to scam will run away.

Contrast with the stock market. There are scams, sure, but the majority of stocks are not scams. If you bought stocks at random you have at least >50% odds of buying stock in a real and honest business of some kind. If you participate in cryptocurrency stuff at random I'd say you have a >90% chance of being scammed... probably closer to 99%.

All the ad-hoc regulations and institutions around conventional finance are the outcome of a multiple century long Red Queen's race between productive economic activity and noise. Noise includes scams, frauds, bubbles, and other forms of undesirable action or emergent behaviors in markets that get in the way of productive economic activity. That hodge-podge of adaptations is almost certainly sub-optimal, but you're not going to do better by just ignoring the problem domain and offloading everything onto individual market participants. That means every market participant has to operate their own SEC and investigative journalist outfit to have any chance of not being scammed. Not going to happen.

A recurrent fallacy of techno-libertarians and techno-utopians is to vastly underestimate human avarice and opportunism. The people designing these systems are usually fairly honest, so they have a hard time imagining what sociopaths and con artists will do with the things they create. If you create anything that offers even the tiniest chance at making money or obtaining some form of status or power, it will be DDOSed with fraud and con artistry of every kind.

Edit: I am not totally bearish on this stuff and I see valid uses for it, but I don't think it's as broadly world changing as its evangelists do.

Exactly. It's the same vicious circle you see with bad neighborhoods, or with community sites that take freedom of expression as their north star. 4chan, for example, was once a noble attempt at a free speech zone, but it has long been known as a wretched hive of scum and villainy. As you say, bad drives out good.
I would also like to address the 'scarcity' claim - when cryptos are being churned out like in the last few years they're hardly scarce anymore ...
Excellent point. The coinmarketcap.com site lists 3882 cryptocurrencies. That makes the "Bitcoin's limited supply means permanent value" talking point ridiculous. They list 91 things with "Bitcoin" in the title, 6 of which are in the top 100 by monetary supply, and all 6 notionally have at least $100m in circulation.
So long as everybody knows what the "real" Bitcoin is, it doesn't present a problem, anymore so than existence of the Liberian dollar renders USD unusable.
It presents a problem for anybody who thinks scarcity will contribute to the value of a digital currency.
Bitcoin is scarce. Applying the term scarcity across multiple cryptocurrencies means you have no idea what you’re talking about.
"In the long term this will lead to novel types of assets, and make ownership extremely liquid. Imagine using your phone to buy shares in a recording artist you just discovered, and selling those shares when they win a grammy."

I'm sorry but I really totally fail to see what is appealing about this future hellscape where literally every aspect of our lives is transacted, monetised and profitable. What is good about this? Why is this something to aim for?

Star Trek was a prophetic vision of the future. Except we aren't meant to be running a federation of planets. We are the Ferengi.
I highly recommend "A Blank Slate of State" by Chris Burniske: https://unchainedpodcast.com/chris-burniske-a-blank-slate-of...

He makes the same argument that I think you're making, that these new types of "more accessible" financial markets in crypto can corrupt people if they only use it to seek out money for money's sake.

TLDR: it's a tool, that can be used for good or for bad, but like any tool it's better to have it than not.

I see the opposite happening, as rent-seeking models like that of the recording industry are replaced.

Also, this regards the music an artist makes and sells, not every aspect of their life.

I don't see how Ethereum or any "digital currency" solve the real world problems that will still exist. Sure, the big players in the recording industry can treat artists badly. But a service is provided, which is doing all business and marking side of things that most artists don't want to bother about.

The idea of recording companies will still exist. Maybe somehow the big incumbents get caught out by the change to a "digital economy", but all that will happen is new ones will appear, and will end up doing exactly the same as the old ones.

Digital distribution was supposed to free artists from record companies. Now it is consolidated to a few large companies like Spotify and Apple because musicians don't want to run and market their own digital store front.

> In the case of Ethereum, digital scarcity secured by a blockchain enables a turing complete state machine that the world can use

This sentence is completely opaque to me. I have no idea what you are trying to say.

> In the most basic terms, this will remove clearing houses for transactions of asset

So I will be able to go to Starbucks, buy a cup of coffee, pay for it with Ethereum and not involve my bank?

> Imagine building a stream of passive income based on the shares you've earned in projects you've worked on throughout your life.

I can currently buy shares in companies I work for. Why is Ethereum different (other than I don't understand it)?

No KYC

Offline account creation

Instantly liquid programmable assets

Programmable banking

Financial censorship resistant

Micro finance

Anyone can use it, it’s surprisingly simple

You could create your own company+shares if you want and they can be instantly available on any market. Ever tried trading a penny stock, or an international stock? It takes forever to open an account at a brokerage where they may or may not allow you to trade certain types of assets, or at certain times. Trades can be reversed and assets confiscated. It’s a million times easier with crypto. Yes it’s a “dark forest” for now (buyer beware) but there are some legitimate platforms like aave and uniswap. Install metamask and check it out! Feels incredibly futuristic to me

> No KYC

I don't know what KYC is but I'm not US based. Sounds like some fee that's peculiar to US banking environment (which I have heard is extraordinarily bureaucratic).

>Instantly liquid programmable assets

Another example of opaque blockchain jargon. Maybe it makes sense, but not as a way to convert the sceptical.

>Programmable banking

I can currently program transfers to happen regularly every month on a certain date, or every tuesday or on soem other timebased trigger. This pretty much covers my personal needs. Everything else I want to double check before approving. I program for a living. Looking around at my family, I don't see anyone capable of writing a simple Python program in order to e.g. pay a bill and my wife has a PhD.

> Financial censorship resistant

Is this a feature? Don't we as a society want a way to control e.g. drug lords and tax evaders moving their money around?

I have never tried buying penny stocks or any type of stock, but is it really such a big deal that you have to wait a day or week before you can place your bets? And that your bets are limited to the roulette and not the blackjack?

> I don't know what KYC is

KYC = Know Your Customer

> but I'm not US based

When sufficiently large sums of money are involved, it's hard to conduct legal business and not play by US rules. Swiss banks discovered this, at considerable expense to themselves and their customers. Cryptocurrency exchanges are going through the same discovery process right now, in both the legal and the philosophical sense.

> > Instantly liquid programmable assets

The unfortunate property of the liquidity of cryptocurrency assets is that they have a tendency to get siphoned away or evaporating in ways not anticipated by their owner.

> > Financial censorship resistant

> Is this a feature? Don't we as a society want a way to control e.g. drug lords and tax evaders moving their money around?

That's precisely one of the questions at stake here. Some folks appear to pine for the benevolent overlordship of the likes of, say, Ross Ulbricht.

>I don't know what KYC is but I'm not US based. Sounds like some fee that's peculiar to US banking environment (which I have heard is extraordinarily bureaucratic).

KYC stands for Know Your Customer, basically anti-money laundering regulations. Banks need to have a decent grasp on whether or not the money they're working with was "earned" legitimately.

>Is this a feature? Don't we as a society want a way to control e.g. drug lords and tax evaders moving their money around?

See above :)

Disclaimer: I think cryptocurrencies are a spillover from a couple decades of inflationary monetary policy from the fed causing people to seek yield in more and more esoteric forms. I also think they're stupid.

KYC = know your customer = anti money-laundering. Yes, it is bureacratic, and it can be a pain for regular, non-money-laundering customers. With crypto, you can instantly create an account offline and trade around the globe.

As far as programmable banking, I meant it in the sense that people can create "bank apps", aka contracts. It goes way beyond simple routing of funds, since it's turing complete.

Financial censorship is a feature, since the finance industry does exclude people and countries. For example, in the US you need to have a $1M net worth or something to be an "accredited investor" which grants you certain privileges. Ethereum doesn't make such distinctions - you can interact with and create any financial product/contract (depending on its rules). Brokerages also regularly halt trading, and make you answer a bunch of questions before trading certain financial products. There are 2 sides to the financial censorship coin so I'm not going to pretend this is always a great feature.

I'm focusing on finance because so far it is the main use case for ethereum. There are a lot of decentralized financial products and derivates now, that even trade in a "USD"-backed crypto (USDC, DAI). Honestly, it all works pretty well, but its not without its downsides and pitfalls.

One of my favorite things about the whole thing is that it gets rid of ACCOUNT CREATION. It's so nice to visit a website/app and just approve it with your wallet. It's the ultimate SSO - there is no email verification crap, its all so instant and frictionless to play with financial products that would otherwise be a bureacratic maze. It's even easier than logging into hacker news.

Anti-money laundering regulations are a good thing to me. I want this, even though it means I had to get a passport made for my 4-year old son so that the bank could see he actually existed and his account total of $50 was not being used by some ukrainian oligarch.

I also think it's ok to put in some speed bumps and force a tiny bit of due diligence before allowing people to gamble away their savings. I'm assuming there is a different risk profile to trading Kazakstani stocks as compared to German stocks, and I think people should be made aware of what they are getting into.

Wirecard and Volkswagen are German companies, you can buy their stock easily.
>Instantly liquid programmable assets

Since nobody really answered this one I'll explain it.

Let's say you create 1,000,000 Blah-Tokens. Problem: your new asset has 0 liquidity. Right? Because there isn't a market for it. If you get it listed somewhere, there still won't be a market until people start placing buy and sell orders. So, can ethereum fix this? Yes! With an automatic market maker like uniswap. Put your 1M fresh tokens in a pool alongside some ETH. We'll say this pool will obey the law xy=k, x and y being the two tokens. We'll also say that the exchange rate shall be x/y.

Now anyone can come along and transact with you and the liquidity is already there! No order book is required. The price is quoted and tokens are added/removed from pool in accordance with xy=k. You get a fee for being nice enough to supply the liquidity. Oh, and anyone can supply their own tokens and split the fees with you. Wa-la.

So liquidity here means whether there is a demand for my Blah-tokens, whether people are willing to buy it for their hard earned cash, gold bars, coffee beans or whatever.

How does Ethereum help create a demand for my Blah-tokens? I can already open a web-shop that sells Blah-tokens. I'm fairly sure the demand will be very low, though. Please explain further how an 'automatic market maker like uniswap' will create demand for my Blah-tokens.

You should probably read up on how automated market makers work. It's going to take a long time to get to the interesting parts if you just guess randomly and expect people to correct you.
It doesn't create demand, it just makes it liquid which would otherwise be a tremendous hurdle.

Liquid markets enable demand though.

What’s the difference between realt.co and REITs?

https://en.m.wikipedia.org/wiki/Real_estate_investment_trust

The difference is presumably the digital contract, that is written in a programming language that only a handful of people on earth can understand, where a programming error can make your money disappear, such that you have no recourse in the court system.

Another difference is the pyramid scheme incentives in the cryptocurrency tech, causing everyone who's bought into it to be incentivised to talk about how great it is.

Please don't comment on subjects you know nothing about and are clearly ignorant and show no desire to actually learn.
I don't really see how something like Realt is any different from owning a REIT. You still have to trust that the property actually exists. AFAIK there is no way for a blockchain to "own" property in the U.S. so there must be some shenanigans going on with a corporation owning it and agreeing to pay out the returns to token holders.

At this point it seems like you've just recreated being a shareholder in a REIT except without all the legal protections.

You can already buy shares of things, crowdfund, buy real estate, invest in REITs, and a million other financial instruments.

Since real estate is a physical thing, it needs to interface with the legal system, the recorder of deeds, etc.

The goal of all this is to circumvent the legal system, but it will never happen for anything physical. Won't even happen for ownership of music groups if anyone ever hopes to use a court.

> In the case of Ethereum, digital scarcity secured by a blockchain enables a turing complete state machine that the world can use.

Does this sentence actually mean anything?

> In the most basic terms, this will remove clearing houses for transactions of assets. In the long term this will lead to novel types of assets, and make ownership extremely liquid.

What does this mean in concrete terms?

> Imagine using your phone to buy shares in a recording artist you just discovered, and selling those shares when they win a grammy. Imagine building a stream of passive income based on the shares you've earned in projects you've worked on throughout your life.

Who wants this?

> ”Does this sentence actually mean anything?”

In more traditional software engineering language, it means “you can run stored procedures on an incredibly slow and limited distributed database.”

Also those stored procedures are immutable, written in a language with more holes than Swiss cheese, and you have to use a scammy cryptocurrency to pay for the privilege of running these unsafe programs. The adherents call this “web 3.0” in some kind of ironic joke because it’s nothing like the web.

> Also those stored procedures are immutable, written in a language with more holes than Swiss cheese...

You're probably referencing this fuck up: https://en.wikipedia.org/wiki/The_DAO_(organization)

The huge advantage of smart contracts is that they are supposed to be trustless - you don't need to necessarily trust your business partners because contract is set in stone ... code and will be executed exactly as agreed beforehand.

In reality, you move your trust in your understanding of the code. But then when 18 000+ (probably highly technical) people investing $50 million don't spot the bug in the code, then it calls the whole "trustless" concept into question.

You don't have to go back that far in time, 'hacks' and bugs in 'smart' contracts happen almost every day. Here's a recent $7M loss: https://cointelegraph.com/news/yield-generating-stablecoin-p... and here's a $89M loss less than a week ago: https://decrypt.co/49657/oracle-exploit-sees-100-million-liq...

It's not even the first time that the 'Compound' smart contract has been 'hacked', but that doesn't stop fools from putting more money into it again.

All these fuck ups are called 'hacks', but actually the code is working exactly as intended, it's just that there are so many ways to write smart contracts incorrectly that it's more than likely they all have hidden bugs, just waiting to be exploited.

"I'd like my investment bank to be an immutable, unfixable piece of code that's a sitting duck target for hackers"

-- nobody ever (except Ethereum ideologists)

Because no code in the financial sector has ever screwed up. Oh wait...

https://en.wikipedia.org/wiki/2010_flash_crash

Oh Oops...

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

Have any ethereum "hacks" ever been prosecuted?

It seems like there would be a reasonable argument that taking value from a contract as defined by code (even if not the intention of the programmer) is not theft.

But would be very interesting to see how a court interprets this.

That doesnt sound like an $89 million loss just unexpected liquidations in loan contracts governing that amount of money
Hmm, this reads like something I do understand, thank you.
Turing completeness in layman's terms means you can write programs on it.

Traditionally a computer that most people are familiar with is a single machine, like PC or the device we can hold in our hands. Etherium is a swarm of millions of computers that anyone can add to the swarm.

When people claim Ethereum is turing complete, it means that anyone can launch a program into it, and the "swarm" runs the program. But which node's output is the truth? Without some mechanism in place, a bad actor could claim to have computed an output that serves them greedily.

Ethereum has baked in protocols so that the distributed result of millions of nodes running a program reaches a stasticial consensus.

Being turing complete means that any computable program can run on Ethereum.

So you can imagine writing a program that, say, runs an auction, or performs escrow for contractors, or whatever. You could run doom on it, albeit at an extremely low framerate because you need the swarm to validate each frame.

I hope this explanation helps.

Edit: I swear autocomplete is becoming more forceful over time.

> When people claim Ethereum is turing complete, it means that anyone can launch a program into it, and the "swarm" runs the program. (...) Ethereum has baked in protocols so that the distributed result of millions of nodes running a program reaches a stasticial consensus.

Usually the reason to run something on more than one computing unit is to achieve additive gains - like getting results faster, or processing more data in the same amount of time. This however sounds like a million computers computing the exact same thing. Why would one want to do that? It's a serious question - I'm having trouble imagining a generic use case.

This is to achieve consensus - the N nodes running the smart contract will ideally replace the central authorities that we implicitly trust.

This enables trustless transactions, which gets rid of a lot of business red tape and unlocks productivity.

Only for a really weird technical definition of "trustless". You have to trust that the smart contract does what's intended, in the way you think it does so. So you need to either trust someone who knows the relevant programming language to audit it, or learn enough to audit it yourself. VS legal contracts, where you have to trust a lawyer to audit the contract but at least have the benefit of laws and bar association ethical rules to constrain the lawyer's behavior. Either way, in practice most people will have to trust someone to safely use any smart contract.
The way I understand smart contracts and them being trustless is that they are effectively an escrow.

So yes, each person participating has to do their own due diligence but once the smart contract is live, it is immutable and therefore trustless since it's just code.

> This however sounds like a million computers computing the exact same thing. Why would one want to do that?

It is what it sounds like, and it isn't ideal. In order for the Ethereum network to reach an agreement on the result of a computation, multiple nodes need to do the computation to check each other. How do you decide how many nodes need to do the computation for it to be accepted by the network? That's not an easy question to answer. The simplest and safest way is to require 100% of nodes to do the computation. Safely getting that % down from 100 to increase the throughput of the network is a problem actively being worked on.

As for a generic use case, think of Ethereum as something like AWS Lambda that happens to have native support for transacting assets.

Can you provide an example where a smart contract needs unbounded loops or unbounded recursion? Do you really need Turing completeness to begin with?
Why does any of this require ethereum?

We have all kinds of markets already, ones that are much faster, more liquid and more efficient than any blockchain-backed exchanges.

There's nothing stopping a recording artist from incorporating and listing shares on any traditional exchanges, or you from making a new exchange.

Literally the only things Ethereum adds are decentralization and no way to dispute or reverse transactions, but for 99% of users those are ANTI-features.

> There's nothing stopping a recording artist from incorporating and listing shares on any traditional exchanges

There absolutely is: it is almost unimaginably expensive and administratively complicated, hence the only entities to IPO these days are companies valued in the tens of billions, who have exhausted all other options for raising capital privately and need to raise at least hundreds of millions or billions more.

I’m no expert on Ethereum or blockchain generally, but the potential for low-friction micro-investing in small businesses, artists or even individuals seems highly compelling to me.

The issue I have with this is that those regulations still exist.

For a recording artist to raise funds in this way still requires that they deal with all the administrative complications - laws just tend to not be enforced very frequently in this new space.

If Jay-Z launched Jay-Zcoin and raised $100M, the SEC and IRS are going to be poking around his finances and looking for any place he violated the law and imposing fines.

I think a lot of the "potential business models" for ethereum that people discuss may be (sometimes) technically robust, but the are also usually extremely fragile politically.

Of course tax laws still apply, as they should.

But Jay-Z raising $100M is the wrong use-case example; he’s already a superstar and already wildly rich so doesn’t need investors, or can easily get them the old fashioned ways (eg record company advance) if he does.

A more apt one is a little-known up-and-comer raising $20k, as they currently might on Kickstarter or Patreon, but possibly using the smart-contract features of Ethereum to link the provision of rewards to some programmed metric like number of Spotify plays.

To trust a currency, you must trust the institutions that manage that currency. You have no control over it.

To trust a cryptocurrency, you must trust the algorithm that runs it. It's fully auditable and predictable.

It still takes me several business days to complete an ACH transaction in the US, and requires I trust the banking system. Transferring cryptocurrencies happen much faster and don't require the same kind of trust or centralized management.

What you see as anti-features are good reasons why they won't replace the dollar yet, but they still have a place in the market.

The US ACH system is extremely slow and still involves nightly batch jobs.

Many other countries have systems that allow for nearly immediate person-to-person money transfer without blockchains: https://en.wikipedia.org/wiki/Faster_Payments_Service This is not a good argument for cryptocurrency.

Regarding "trusting" an algorithm: in the real world you also have to trust the implementation, and the configuration, and eventually humans involved in the transaction. How often do we see some big data breach on HN? Very rarely has any cryptographic algorithm been broken, but often one of the other links in the trust chain has broken.

> Very rarely has any cryptographic algorithm been broken, but often one of the other links in the trust chain has broken.

The whole point of Ethereum is to reduce/eliminate those links in the trust chain.

>To trust a cryptocurrency, you must trust the algorithm that runs it. It's fully auditable and predictable.

What happens when Ethereum 2 inevitably hits a security bug, much like the ones Ethereum 1 hit that led to the DAO hack/theft (some might say... worked exactly as coded)?

Compare it to traditional security. If you want to steal a few dollars you just need a sharp bit of metal and an easily frightened shopkeeper.

Of course when you leave, the shopkeeper (assuming you didn't murder them) can use a phone to inform the authorities. Then a manhunt begins involving resources like police and their cars, radios, guns, protocols, legal justice system, forensic investigators etc.

Is the sharp piece of metal a bug? Is the patch just more force, monitoring and authority?

Edit: And is the cost and risk of trusting an algorithm comparable to the human cost and risk in maintaining current securities "authenticity", which is basically a might is right system?

>We have all kinds of markets already, ones that are much faster, more liquid and more efficient than any blockchain-backed exchanges.

You are talking probably from a developed country point of view. For those like me who lives in a developing country none of those are accessible for everyone, only for a few percentage of the society. Cryptocurrencies solves a lot of problems for people in oppressed countries or with a lot of terrible regulations and high inflation rates.