4 ms·
Here are some general thoughts for those interested more in this topic: - Ethereum and Bitcoin both let you lock up the cryptocurrency and provide a series of
by nemild 10y ago
Here are some general thoughts for those interested more in this topic:
- Ethereum and Bitcoin both let you lock up the cryptocurrency and provide a series of rules that are required before the money can be unlocked; this has a lot of power, because unlocking can be done by a single party (a standard payment use case), but also by a series of other people/groups or external events (currency price, weather, etc)
- The rules for unlocking cover a wide gamut from requiring several users to agree (e.g., multisig like a multi key safe deposit box to voting for a dividend), to requiring an external action to happen (e.g., unlocking if the temp is below a certain amount on a given day for weather based insurance contract); these actions replicate many current real world interactions like an insurance contract, a voting mechanism for corporate governance, and escrow (just to name a few); the OP has also written about Distributed Autonomous Orgs (DAOs) on TechCrunch
- A key benefit for letting these be programmatic is that they can cut out a middle party (or at least reduce the dependence on this middle party with a network of oracles) - cutting costs and reducing dependence on this third party
- Some key value props where this could help in the early days:
1) cross-country transactions, as the contract language is international (legal systems can be hard to enforce across geographies)
2) Small amount transactions (as fixed costs may be high in a real world contract today for each party)
3) contracts where middlemen have monopoly-based pricing
4) contracts where the middle parties have a low degree of trust (more likely today, certain markets don't exist due to this reason)
5) Contracts in a country where legal systems are weak/expensive
- For Bitcoin vs. Ethereum, many Bitcoin enthusiasts will argue that Bitcoin provides a similar functionality, but Ethereum's high level languages are generally much easier to write (I'll likely get some push back by partisans, but Ethereum's Serpent and Solidity might be closer to JS/Python/C, while Bitcoin's Forth-inspired/stack-based language is closer to a higher-level assembly); there are also some specific differences like Bitcoin requires all or nothing unlocking, is not turing complete, has restrictions on the transaction byte size, and has some miners enforce transactions that only have a few of the contract op codes, not all; still, some Bitcoin devs (like Gavin) note that the surface area for security holes is significantly larger with Ethereum
(See my other post for an introduction to the language)
- daoland 10y agoOne key difference is Ethereum is not being built only for writing code for monetary transactions using Ether. Ethereum is being built to store and run code in the EVM. Ether main use is only to pay for gas so that any user of the network can execute the code. Although in Bitcoin there is no point in writing code other than encode conditions upon which monetary transfers can be done, this isn't Etheruem's main goal. An example of a non-monetary use case is - A concert ticket issuer could issue tickets on the Ethereum network. When users want to resell, they need not rely on a re-seller and instead can do a direct exchange on a decentralized exchange. They will pay a few cents worth of Ether as fees to run code that triggers this exchange to get the guarantee that the tickets are not counterfeit. After a quick confirmation the ownership of the tickets is updated on the blockchain. The exchange of money between these users can be in USD itself directly without any conversions to Ether. It is non-monetary consumer use cases like these that drives the Ethereum team to bring the transaction confirmation times to the current 15 seconds and even lower in the near future.
- DennisP 10y agoIn practice I wouldn't say that's ether's main use. If you add up the total gas fees so far, it'll probably amount to less than the $150 million in ether raised for TheDAO, the $5.5 million raised for Digix, etc. And if you look at the code samples on ethereum.org, one is for a crowdfunding app that raises funds in ether. And of course, if you're paying miners with ether, then ether better be a functioning currency or it's a worthless payment. You're certainly right that Ethereum has lots of uses beyond simple value transfer, but if you want to transfer USD instead of ether, you need to trust someone to back tokens with USD, and trust governments not to shut them down over KYC issues.
- daoland 10y agoThe devs are so badass, they constantly mention that they will allow miners to accept any currency for gas fees if it is feasible. Most of us who have some Ether are comfortable with this move too, ironically our hedge for this scenario is to hold some Bitcoin. Because, Ethereum will prosper not when Ether has a high market cap, only if it can become robust enough to provide a decentralized platform to run code for a variety of apps and services.