Hacker News Reader

Comment by nostrademons | Hacker News Reader

Interesting to read this with an eye on the authors' mindset. Their understanding of stablecoins seems largely centered on Tether (and to a lesser extent, BUSD/USDC). A lot of their understanding is incorrect when applied to algorithmic stablecoins like Dai, eg. there is no central issuing authority; Dai is minted in exchange for Ethereum (and other cryptocurrencies), not fiat currencies; the effect of a run on Dai is unlikely to spill over into the mainstream financial system; a "custodial wallet" in DeFi is not a company but just an Ethereum address, i.e. a hash of a public/private keypair.

Overall I get the sense that the government is still people/organization centric and cannot wrap its head around a future where reality is determined by computer code and people are bit players in the script.

View on HN

From the footnotes:

"""Stablecoins that are purportedly convertible for an underlying fiat currency are distinct from a smaller subset of stablecoin arrangements that use other means to attempt to stabilize the price of the instrument (sometimes referred to as “synthetic” or “algorithmic” stablecoins) or are convertible for other assets. Because of their more widespread adoption, this discussion focuses on stablecoins that are convertible for fiat currency."""

Ah, interesting. Dai is mentioned several times within the body of the report, though, so it's implied that it's covered within the scope of the report.

I'm sure that the distinction will be lost on whatever press cycle or legislative output this report generates.

Perhaps I'm oversimplifying your comment, but it seems like a very good thing that the government continues to be people/organization centric and doesn't embrace a future where "reality is determined by computer code and people are a bit players". We should hope our democratic institutions continue to operate this way.

I'm value-neutral in this comment thread, I'm just describing a.) how things are and b.) how the authors of this report are making assumptions about how things are. I can see plusses and minuses for both people-centric and code-centric approaches.

I get that it's pretty natural that people would consider it a bad thing for people to hand over power to computer code. It makes perfect sense if you consider yourself not as a person but as a collection of electrical impulses floating around in your brain, though. Computers are the same thing, they just transmit those electrical impulses millions of times faster.

Why are you so committed to erasing the difference between humans and computers in order to place humans below computers?

Realistically I don't have the power to put humans either above or below computers. That'll depend upon the independent choices of billions of humans and hundreds of billions of microprocessors.

I've found that having an accurate model of the world - one that can generate likely predictions - is very important for profit now and potentially for survival in the near future. Knowing that people are frequently blind to changes in the world that do not place them at the center of the universe, it's worth correcting for that bias in myself and envisioning a world where humans are not necessarily on top. What would that look like, and what's my best chance for survival in such a world, given that I am human?

Sure, but your comment doesn't reflect an accurate model of the world -- brains and computers are not the same kind of thing, and humans are more than just a brain. I know there are powerful people trying to convince us that your model is accurate, though, with the goal of getting us to allow their computers more control over our lives. I'm worried that people are buying into this view, not because they believe it's an accurate model of the world, but because they think it's inevitable that these powerful people and their computers actually will end up in control, so they'd better prepare for that future.

Everything is already under control of three companies world wide. Vanguard, Blackrock and Berkshire Hathaway own everything worth owning. Who owns them?

Computers are a tool built to improve the lives of people. It does not make sense to consider them an end in-and-of-themselves.

So is capitalism, and yet capitalism has seemingly taken over human behavior in a way that many people find is not an improvement, and yet are powerless to stop.

Complex systems often exhibit complex emergent behavior. Humans are part of that system. Humans are not the only part of that system, and it makes sense that agents interacting at speeds millions of times greater than us might eventually come to dominate the system.

> capitalism has seemingly taken over human behavior in a way that many people find is not an improvement, and yet are powerless to stop.

Lots of people banded together to rollback the degree to which capitalism dominanted life in the developed world, and were pretty successful at it over the last century.

> Computers are the same thing

Uh, no? Computers aren't people.

You have a valid point, and to some extent government should continue to be people centric. But there is value in replacing a some of the system in place by code.

IMO, a lot of ambiguous laws and polices create different outcomes for similar inputs, and that should not be the case of legal systems and institutions. Money is perhaps the first of the government aspects to be easier to express in decentralized code, but I won't be surprised to see this as only the beginning of a trend. Think autonomous cars replacing drivers and other interesting similar developments.

I think they approached it from two directions:

One is that they broke down stablecoins into the following activities:

* Governance

* Management of Reserve Assets

* Custody of Reserve Assets

* Settlement

* Distribution

Just because one, more, or all of those functions are managed by smart contracts does not mean that the others cannot be regulated. Regulation may be the requirement to have regular contract audits for example and, in the case of Dai, the Maker Foundation would be responsible for adhering to those regulations.

Second, I don't think the writers of this paper perceive nearly as large of a risk from algorithmic stablecoins as they do institutions which claim to maintain asset backing in the normal financial system. The former have the transparency of the chain as a backing, and that transparency makes the currency peg safer. The latter have no transparency and regularly seek to obscure, which makes them extremely dangerous as they grow in scope.

DAI has a market cap of ~$6.5b and the transactions happen on chain. USDT has a market cap of ~$70.3b and has 0 transparency. It is clear where regulation should be focused.

> there is no central issuing authority; Dai is minted in exchange for Ethereum (and other cryptocurrencies), not fiat currencies

Its not fiat-stable, which is probably their focus. fiat-stable coins are basically crypto bank notes:

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

Also,

> a future where reality is determined by computer code and people are but players in the script.

Corporations are "things" in a legal sense - but they're still managed by people, created by people and owned by people. DEFI contracts and orgs are still made by people, and sometimes also managed by people. You can probably be held liable and tied to your misbehaving contract.

The 9th Circuit ruled that source code is protected under the First Amendment in the Bernstein case, which legalized the export of cryptography. If you just publish the contract and don't have any ongoing administration, I would think that gives you a pretty strong legal defense.

Publish the code, or activate/enable/fund/start/whatever to instantiate the activity.

If you just publish it github - thats probably a good defense (linux T isn't to blame for all bad things that ran on linux).

If you publish a smart contract in a way that enables people to start using it, then that may not be covered. While its different than actively running an API/Service that needs continuous/paid hosting, its probably a grey area at best.

If the government were to take a hard line on this, then publishing on github for anyone to deploy would work. It's easy to verify that the compiled contract matches published source code.

But the Bernstein case was about PGP, and after the decision people had no trouble with using it, making compiled code available for download, etc.

> Overall I get the sense that the government is still people/organization centric and cannot wrap its head around a future where reality is determined by computer code and people are bit players in the script.

Conversely, people in government tend to view crypto folks as senselessly computer-centric and unable to view macroeconomics as the result of human interaction.

There is no such thing as value without exchange. That's a definitional thing. Coins themselves are just numbers.

Exchange value only matters when there's trade. My view of crypto is "wrong in the right direction" because it currently addresses macroeconomic coordination via mimicry of traditional finance. The technologies it uses to achieve decentralized coordination do matter, though. Automated consensus on a large scale is meaningful. But it needs a post-trade, natural-systems view of the world to really make sense.

I mean what's the point of that other than to serve the very small number of people who own the resources to edit 'the computer code'?

The internet and computers, despite being very complicated tools, are still just made by humans to serve humans.

I suggest you ask yourself which humans want the outcome you described and why.

DAI is not considered an algorithmic stablecoin, but an asset backed stablecoin. Algorithmic stablecoins like FEI do not seem to work as well as asset backed stablecoins.

Wait, why is DAI not algorithmic?

> a future where reality is determined by computer code and people are bit players in the script.

As opposed to a reality determined by physics? Perhaps you're referencing AI overlords? I suggest that you may not understand the purpose of states. Even algorithms are the expressions of people.

Perhaps you can explain what you meant in a way that I'll more easily understand?

> a future where reality is determined by computer code and people are bit players in the script.

When you put it like that, it does sound a bit creepy.

It is the government's job to translate abstract concepts to be people centric, simply because the world itself is people centric. Digital bits have no power unless they is a shared recognition of that power, same as a written contract, a book of laws, an election or anything else.

That computer code runs somewhere and every IP address leads to a person eventually.

Except it doesn't. On Ethereum and other blockchains, that computer code runs everywhere, and every IP address is a gossip protocol that may have originated an undetermined number of hops backwards.

The lifecycle of an autonomous program (colloquially called smart contract) is that it is deployed by an address of a human being or that human being's server, and then the address' first transaction to the autonomous program is to delete the address' administrative capabilities of that program. Autonomous programs live on every validating node of that blockchain, and those validating nodes have no knowledge of the behavior of those programs. The deployment feature publishes the code on all validating nodes, no different than any other transaction. All future behavior of the autonomous program comes from individual users who do not control it.

So for what grandparent poster was referring to, stablecoins collateralized by digital assets, all the collateral is provided by users and all the stablecoins issued were caused by users providing collateral. Those users clearly do not run the autonomous program, no different than a depositor at a bank is not responsible for the bank when they ask for a loan from the bank. There is nobody to sanction, and there is no way to disable the autonomous program that accepts collateral and issues collateralized stablecoin loans.

Also, within EVMs (a type of development platform, growing category of blockchains), the users do not have a record of an IP address (although the node they connect to can record it, to mitigate that the user can run a relaying node from their personal computer. relaying nodes forward to validating nodes. no nodes in an EVM have knowledge of other nodes IP address and no nodes are even aware of which node saw a transaction first). And regarding the tracing of their onchain address, a user can provide collateral from a virgin address funded by other autonomous programs like Tornado which sufficiently mix funds. The programs and the regulators are not capable of factoring in our opinion about that.

Legislation will ensure that reality continues to be determined by legislation, not computer code.

Reply on news.ycombinator.com