Flash Loans Are Providing Instant Cash to Crypto Speculators
bloomberg.com
bloomberg.com
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.
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.
The real enemy is the closed fractional reserve banking system, I think practucally any open source cryptocurrency is better than that.
~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.
Try asking $100M for a short time from any bank, I'll see how successful you are.
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.
Nanoseconds of computational time.
If the trade isn't possible then the transaction fails and it doesn't go through at all. You pay the transaction fee.
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.
you still need to know what markets and contracts you are targeting and with that variables
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
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.
https://www.bloomberg.com/opinion/articles/2021-02-05/robinh...
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.
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.
We will have some crypto winters in the meantime, but the trajectory is undoubtedly going upwards.
Since then, the institutionalization + litany of new crypto to generate wealth for their originators has really soured that feeling for me.
That said I still thing it's gonna be a chunk of the future of human civilization somehow.
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.
[1] - https://cryptofees.info/