Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ethereum: Rise of the World Computer

svds.com

Post reply on HN
Print view View original

71–80 of 243 posts

Re: Ethereum: Rise of the World Computer

#71

Post by jsprogrammer »

pjc50 wrote: post #27
teraflop wrote: post #13

The article is pretty light on technical details. Does Ethereum suffer from the same design limitation as Bitcoin, where every operation must be verified by every node? If so, I have a really hard time believing it will scale. Look how much difficulty the Bitcoin network is having with just a couple of transactions per second.

Isn't that inherent to blockchains (every node must have a copy of all transactions)?

You don't need to keep all the transactions. The original paper addresses how to compact the chain. Basically, you only need to keep hashes.

Re: Ethereum: Rise of the World Computer

#72

Post by crazydoggers »

kang wrote: post #52

Previous discussion https://news.ycombinator.com/item?id=9987142 Give me any ethereum use-case & I'll tell you how to do it with bitcoin. Is it advantageous if a DSL is turing complete? It is not that bitcoin 'cannot have' a turing complete script but that it is unsafe & there is no need to do so. Turing completeness can be applied on an upper abstraction. There is nothing that ethereum can do that bitcoin cannot! Th…

> Give me any ethereum use-case & I'll tell you how to do it with bitcoin

That's like saying give me any HTTP use-case, and I'll tell you how to do it with Gopher. To me Rootstock is like putting the WWW on top of Gopher. You can do it, but why?

"We don't need GUIs, command lines can do everything!"

"We don't need Java/Python/Ruby, you can do it all in C!"

"We don't need C, you can do it all Assembly!"

Ease of development and implementation really matter. If I can accomplish creating my block chain app on Ethereum in a fraction of the time and complexity of the same thing on Bitcoin, I'd have to have an extremely compelling reason to choose Bitcoin instead.

Re: Ethereum: Rise of the World Computer

#73

Post by jcfrei »

I've dabbled with the dev tools ethereum provided for a while and I can't imagine that the computations ethereum does will ever be fast enough for a general purpose, distributed computation platform. If I had to wager on yet another cryptocurrency™ I'd bet on Zcash (https://github.com/zcash/zcash). They promise a real differentiator for cryptocurrencies with their zero-knowledge proofs.

Re: Ethereum: Rise of the World Computer

#74

Post by desdiv »

kang wrote: post #52

Previous discussion https://news.ycombinator.com/item?id=9987142 Give me any ethereum use-case & I'll tell you how to do it with bitcoin. Is it advantageous if a DSL is turing complete? It is not that bitcoin 'cannot have' a turing complete script but that it is unsafe & there is no need to do so. Turing completeness can be applied on an upper abstraction. There is nothing that ethereum can do that bitcoin cannot! Th…

>Give me any ethereum use-case & I'll tell you how to do it with bitcoin.

How do you implement a trustless SatoshiDice with bitcoin?

Re: Ethereum: Rise of the World Computer

#75

Post by woah »

Uptrenda wrote: post #51

Crypto has come a long way since Ethereum was first conceived and what was impossible only a few months ago is now 100% practical. Today there are A LOT of new techniques that allow us to largely implement decentralized autonomous applications on any blockchain (no matter how limited) - a fact I feel often gets missed out in all of the Ethereum hype. If you have a lot a techniques like zero-knowledge proofs, homomorp…

I'm trying to figure out your angle here. Are you saying that ethereum isn't ambitious enough because it doesn't incorporate these fun crypto techs you mention? I would say that it's already an extremely ambitious project so it's good that they didn't bite off more than they can chew.

Or maybe you are trying to say that this could have been done on Bitcoin? Maybe, but whatever you try to shoehorn into Bitcoin's crippled contract pseudolang is going to be an order of magnitude simpler in a language that is Turing complete.

Re: Ethereum: Rise of the World Computer

#76

Post by kang »

jarsin wrote: post #70
kang wrote: post #65

Earlier quoted context omitted.

Please give me an ethereum use case & I'll try to tell you how you can do it with bitcoin. I have previously answered a few use cases here [ https://news.ycombinator.com/item?id=10846788 ]

the bitshares, nxt, insert any blockchain project all say the same thing yet none of them comes even close to the amount of dev taking place on ethereum. hmm why is that....

Vitalik became famous cause of his journalism. Then the sale gave ether founders loads of money. Now it is being used to market cleverly. One example around me here in India is BlockchainU. Blockchain university has taught in almost all prime Indian IT companies pumping ethereum & thus atleast most, if not all, corporate devs are working on ethereum here, whereas there is no central advertising foundation for bitcoin.

Re: Ethereum: Rise of the World Computer

#77

Post by mbilker »

na85 wrote: post #46

Has it been six months since the last Ethereum post already? Is Ethereum still "engendering positive disruption"? I see they've updated their site to make it more approachable and read less like proecdurally-generated Venture Capitalist jargon. That's nice. I see they've added a javascript environment. Yuck.

They had the javascript environment for awhile now.

Re: Ethereum: Rise of the World Computer

#78

Post by kang »

desdiv wrote: post #74
kang wrote: post #52

Previous discussion https://news.ycombinator.com/item?id=9987142 Give me any ethereum use-case & I'll tell you how to do it with bitcoin. Is it advantageous if a DSL is turing complete? It is not that bitcoin 'cannot have' a turing complete script but that it is unsafe & there is no need to do so. Turing completeness can be applied on an upper abstraction. There is nothing that ethereum can do that bitcoin cannot! Th…

>Give me any ethereum use-case & I'll tell you how to do it with bitcoin. How do you implement a trustless SatoshiDice with bitcoin?

You can do it using the appcoin concept [http://startupboy.com/2014/03/09/the-bitcoin-model-for-crowd...]

Re: Ethereum: Rise of the World Computer

#79

Post by pipermerriam »

jcfrei wrote: post #73

I've dabbled with the dev tools ethereum provided for a while and I can't imagine that the computations ethereum does will ever be fast enough for a general purpose, distributed computation platform. If I had to wager on yet another cryptocurrency™ I'd bet on Zcash ( https://github.com/zcash/zcash ). They promise a real differentiator for cryptocurrencies with their zero-knowledge proofs.

Have you read up on the proposals and information on plans for Proof-of-Stake (Casper) and Scaling/Concurrency via sharding? Both of those are directed at fixing the "fast enough" problem.

Re: Ethereum: Rise of the World Computer

#80

Post by kefka »

cjm wrote: post #67
kefka wrote: post #62

Unfortunately, I'm trying to obtain a small amount of ethereum currency to experiment with this.. And I just killed my thread doing 1M hashes a sec, because I found out that 6MHash/sec = 5 Eth/month on average. I'll spend my time learning other things, mainly because this is inaccessible to me.

You can buy the Ether on an exchange.

I was hoping for a roughly low barrier of entry, rather than try to figure out who will trade USB for ETH, as I have no BTC.

In all honesty, I'm looking at distributed applications as well, which from what I read Ethereum says it does and doesn't do. It's a bit fast and loose from what I read about, about Ethereum's capabilities.

I'll go back to my experimenting with TOR and how it might be possible to make your own distributed applications within the Hidden Services. That's free to use, and as low bar as I can think of. (Add machine, shows up as HS.)

Post reply on HN