Flash loan is just money that an Ethereum contract gives to a contract given in a callback function and expects the money back inside the same transaction with an interest. It's the first loan with no counterparty risk.
While I'm not a fan of how the Ethereum project and ETH token is managed, Solidity programming language brought real financancial innovations in trustless contracts.
> While I'm not a fan of how the Ethereum project and ETH token is managed
There is a real possibility of BTC finding home on Ethereum network in future. Recent advancements in cryptography make building of fully trust-less bridge between two possible, without requiring any changes on Bitcoin side. So possibility will be there. As for the need - there is no indication of Bitcoin moving towards solving time-bandit attacks. They will become a real issue once block reward falls. I'm skeptical that Bitcoin as a project is even capable of solving this problem. It would require additional emission of BTC, and "21M BTC meme" is way too strong.
Sure, we can talk about this in 100 years when BTC emission stops. Right now I prefer staying in BTC, but I think that the 2 big networks will grow together.
The real enemy is the closed fractional reserve banking system, I think practucally any open source cryptocurrency is better than that.
Check out some of the V3 crypto's. DOT, ATOM, ADA.
Most of the OG founders who were building Ethereum realized sharding wasn't going to work and left to build the above three.
I recommend to stay diversified among crypto this decade. And if you buy DOT on Kraken it gives 12% more DOT per year.
Pump it. There is very little difference between DOT and ETH 2.0. They are both sharded classical consensus. ATOM is also classical consensus and once the IBC will be up it will also resemble sharding. If you really want to look at something different in terms of consensus than Avalanche is something to check out. They claim they have introduced 3rd category in consensus protocols, but we will have to wait and see if it stands up to scrutiny and real life usage.
> Sure, we can talk about this in 100 years when BTC emission stops.
~89% of bitcoin have already been issued. Yes, they will continue to be emitted for a long time, but at a rate that is rapidly approaching inconsequential.
It’s funny that it’s too small for you, while others complain that it uses too much energy. I guess it can’t please everyone, but for me it’s perfect how it is right now.
Ask redditors on Wall street bets: when the lending party has counterparty risk, the rules of lending are opaque and can disappear in a second without a clear explanation.
Try asking $100M for a short time from any bank, I'll see how successful you are.
What is the liquidity of ETH lending market? 100M short term loan easy to get, what would be the interest?
Last time I checked defi usdt lending rates seemed to hover around 8-9%. Not that exciting when you consider insane gas fees and all the hassle managing the contracts.
I've been researching flash loans for a whole year and its still clear to me that some market participants have a better way of analyzing all markets (the contract state of smart contracts hosting markets) than I do. I feel like they also image the current state of the network and execute variations on Ganache/localhost before sending it on the live network.
Furucombo is the only thing I'm aware of that only partially helps in the structuring of flash loans, but even then using their contract is too expensive.
Over the past year I've built oracles and deployed fleets of complex smart contracts, learning about many areas of improvement in smart contract land and seeing a lot of vulnerabilities in other team's contracts, but as far as actually doing anything with that I'm still very far off.
I'm far from an expert on this, you probably know better than I do, but wouldn't forking the blockchain at the block height you're interested in with Ganache and execute it there be enough? or is the issue the transaction queue?
I'm curious about exploring this myself, would you mind explaining in more detail what is lacking after forking the main net with ganache?
Is it that you need a high powered server to fork it continuously, and that smart contracts tend to cost a lot to execute, or is there more to it than that? I don't understand what contracts and variables that you're targeting won't be included with the solution we're discussing
You need an archival node to store all the states and also monitor unconfirmed transactions. The primary client go-ethereum (geth) creates archival nodes that grow 1 terabyte per year (7tb now).
Despite archival nodes not being needed for consensus, some people’s kneejerk response is that this means Ethereum is not sustainable, but really geth just has unoptimal coding.
There are other clients out there that have chosen different data structures and reduced this by an order of magnitude. Turbo-geth, for example can sync an archival nodes at 1tb and is working on other improvements.
There is more support for geth at the moment but this is also an area of development if you want to improve upon.
Its likely many nodes out there are using their own optimizations and are not offered to the community.
It gets even funnier when you consider that miners may reorder transactions they include in a block as they see fit, and even add their own transactions to front-run others.
What is the objection? Fees are designed to entice miners to include a transaction. If a fee is too low, the miner will include transactions with better incentives. Recording transactions has real economic cost (mining and permanent storage on the distributed ledger).
I was about to reply that the "front-running" has been widely reported, decided I'd better google a source, and came across a fantastic article about how payment-for-order-flow actually works. You're right, it's not front-running, and it gives retail traders a better price than they would have gotten on the market itself.
Recently miners started exploiting this possibility for profit. This is a problem in many ways, and this problem is hard to solve. Possible solution involves splitting miner role into two - one group will do sequencing (without knowing much about txes contents), other will do application of txes to the state managed by the chain. There are more radical ways of solving the problem - for instance one can hide all the execution from miners. One example would be ZEXE: https://par.nsf.gov/servlets/purl/10175111
Crypto was great at feeling superior than the financial world when BTC was simple and pure. Now we have earning interest, rehypothetication, DeFi and a whole lot of weird stuff that no one understands. Plus its still useless for actually buying things.
Most of the issues with complex financial products leading to the 2008 Global Financial Crisis was the lack of pricing and provenance.
People did not know who owns what, what the underlying assets were, and what the prices were.
In a system where all states are visible and the prior states unchangeable, these cease to be issues, allowing for many of those complex products to exist very easily, which is exactly what is happening and has happened.
If you don't trust that, there is also smart contract insurance you can buy for when smart contracts and oracles create results outside the bounds of the community understanding.
This doesn't mean there won't be problems, an insurance smart contract recently got hacked, it is prudent to recognize that the solutions are presented and the market is choosing those solutions. You should be in the place to laugh about it and know what specific issue occurred and know how you can improve that, as opposed to simply waiting for me to finish so you can play devil's advocate on a mere concept of "crypto" because you might be stuck in the wrong decade since we're way past that.
I'm okay with the rapid iteration towards stronger more robust systems.
The permissionless nature is driving that rapid iteration. Unlike the outsiders in "The Big Short", you don't even need an ISDA Master Agreement to trade these esoteric products. And the flash loan market is very similar to the Federal Reserve's overnight repo market, you don't have to be a big bank. In the repo market, the Fed creates and destroys hundreds of billions of dollars for minute long trades on request, allowing banks to complete deals without needing the cash themselves. This is very onbrand to what "crypto" aspired to do, lower barriers of entry. The permissionless nature is specifically that it doesn't require your opinion or any gatekeeper in order to offer something to the market.
It doesn't solve the problem, most of the issues are related to current prices with oracles instead of not knowing any price. Services in the DeFi space introduce solutions to prior iterations of complex products hiding the risk.
They aren’t trying to show how its better than traditional wall street, there is no one person or organization saying “eureka now we can offer the biggest CDOs since Lehman”, its new people picking up new tools building what the community wants, or else they dont have a community.
I feel the opposite: defi has been shaping up to be seriously useful. There's still a lot of kinks to work out and while I don't see it replacing tradfin systems, the next 5-10 years look very bright!
We will have some crypto winters in the meantime, but the trajectory is undoubtedly going upwards.
I first became aware of BTC around 2011, bought some around 2013. I didn't really use it for buying anything at all tbh, but it still felt fun - there was a playful, toy like element to it (for me).
Since then, the institutionalization + litany of new crypto to generate wealth for their originators has really soured that feeling for me.
Maybe BTC succeeded at providing people with free (as in freedom) money, so some of it is inevitably going to all sorts of unregulated uncensorable trading games.
Does anyone have a good book, or resource to actually understand this stuff? Seems like the information is spread across many different projects and every book i buy focuses on blockchain development.
What do flash loans do that normal leverage doesn't? Am I correct in wondering whether this allows the unskilled or unlucky trader to ruin themselves a little more quickly?
If the transaction doesn't immediately (within the same block) result in enough profit to pay back the loan with interest, the loan is canceled altogether, i.e. basically it doesn't even happen. Then you only pay some gas fees for executing the smart contracts up to that point (which can be a few hundred bucks nowadays, I must say).
More than that. You can actually simulate your execution locally. So you should know if there is a possibility of making a profit even before you send the tx.
Is there any reason why Uniswap would use something like Ethereum over Algorand (for example)? The latter seems to offer very low tx fees and much higher tx rates, is it a matter of smart contract code etc?
> The latter seems to offer very low tx fees and much higher tx rates, is it a matter of smart contract code etc?
Chains are of two types. Cheap and unused is one, expensive and used is the other. For Ethereum current limits on the size of the block (and thus tx fees and tx rates) are set because of lack of solution for fast state size growth. Algorand and every other "ethereum killers" also don't offer solutions for this problem - they don't have it because nobody is using them. Ethereum has few candidate solutions on the table. The major one is "stateless client" - solution where tx will would contain commitments for every bit of state which is read or modified. Basically paying with bandwidth to decrease the size of actively managed state.
There's many solutions coming online this year. Polkadot is launching a parachain where you can execute your eth contracts using the polkadot network. Also many L2 solutions will solve the problem as it is just a problem of 'Compute'. The transactions gain some 'Centralization' but that doesn't seem to matter in the industry anymore. It's more about being 'Censorship resistant'.
While I'm not a fan of how the Ethereum project and ETH token is managed, Solidity programming language brought real financancial innovations in trustless contracts.