4 ms·
It will burst when everyone who could possibly be convinced to buy has bought. When the optimism is so contagious you start to doubt if it will ever really pop,
by rthomas6 9y ago
It will burst when everyone who could possibly be convinced to buy has bought. When the optimism is so contagious you start to doubt if it will ever really pop, that maybe this time it's different.
In the dotcom bubble, people were buying shares at prices such that justifying those prices would require companies to make more money than any company had made before. But people, including experts, said this time it's different because the internet will change everything, and those companies really will make more money than any company has before. Even though no actual hard data justified that optimism.
- charlesdm 9y agoI highly doubt we're close to that point, to be honest. This is a global bubble. Billions of people around the world. Most of the money invested will flow in a few top coins -- not 200 different equities. The dot-com bubble was US centric, and the market went up to $6.7tn before imploding to $1.7tn. If, in the mania of the 2000s, things could head towards $6.7tn, I don't believe that this bubble will be smaller in the age of loose monetary policy. Money is essentially free today. A few weeks ago, Saudi Arabia paid $450m for a painting. If that isn't a sign of how money has lost its value, I don't know what is. Would I bet everything I own on crypto? No, of course not. Can you still get a 10x return on money you invest today? I wouldn't be surprised. Even knowing there will be a major correction at some point, it's probably a mistake not to own at least some.
- rthomas6 9y agoYeah, I own some ETH and EOS (not very much), because those are the ones I most believe have long-term intrinsic value for the world.
- stuxnet79 9y agoCan you summarize what the differences are between ETH and EOS?
- rthomas6 9y agoSure. They're both Turing-complete blockchains which can run decentralized apps/smart contracts, and of course be used for transactions directly. Ethereum (ETH): Created in 2015, currently successfully running. Right now ETH uses a proof of work consensus algorithm with plans to move to a proof of stake algorithm over time. Its smart contracts can be coded in two purpose-built languages, and work is being done on cross-compilers to facilitate writing in more. ETH can currently handle ~20 tx/sec, but the community is very active in working toward vast improvements in this regard via very technical solutions (sharding, plasma), but they're not quite there yet. Ethereum is truly decentralized and runs tens of thousands of validating nodes. EOS: Planned creation date is in June 2018. Its smart contracts will use WASM (allowing the use of many general purpose computer languages), support parallel computation, and have some sort of ability to change deployed contracts to allow for debugging and updates. It will handle 100,000+ tx/sec out of the box, and transactions will have 0 fees. Instead of fees, percentage of holdings determines allowable transaction bandwidth (If you have 1% of EOS, you get 1% of EOS transaction throughput). However, instead of tens of thousands of validating nodes, EOS will have 21, with all coin holders having the ability to vote for who these validators are. EOS will trust the consensus of these validators for transaction security. This is how EOS achieves such high transaction volume and zero fees. IMO, this is the main weakness of EOS.
- stuxnet79 9y agoThanks for the breakdown. Very helpful. I'm somewhat disappointed with EOS's offering compared to ETH, as I'd always (wrongly) assumed they were providing a lot of innovations in an effort to reduce ETH market share in the whole 'dApp platform' space.
- rthomas6 9y agoI think they do in some regards. 0 fees to run dApps combined with WASM-nativeness (write in many languages) seems objectively better than ETH when considering just those areas.