> Hasn't the Electrum model worked for Bitcoin?
Yes, it has. But I describe in my whitepaper why it doesn't work for protocols on top of Bitcoin.
> Is there a big issue with just querying a bunch of supernodes to see if they all give the same result?
That is potentially very problematic because it breaks down the no-trust property; instead, you're getting data from a small pool of centralized entities. It's possible, and people do it with blockchain.info, but it somewhat weakens the whole attraction of a cryptocurrency.
> How many years from now should we predict that the blockchain size will become irrelevant?
That's an interesting question that I don't know the answer to. Perhaps if that happens the blockchain size will simply keep on growing, as people keep putting more and more memory-intensive applications on top and reach an equilibrium. In the medium term, I personally am more worried about the possibility that Ethereum's disk and computational load will be too high; I think that is the more pressing question that we at the Ethereum team are more focused on.
> What will be the effects of ethereum blockchain size growth several orders of magnitude greater than Bitcoin?
I don't think the blockchain will be that large. There are ways to combine Ethereum with off-chain protocols to mitigate most of the issues if people do want to put terabytes of data into Ethereum-based systems.
> Is it reasonable to think the client can be sandboxed?
> What will be the effects of illegal contracts? An assassination market, for example? Will the Silk Road be able to run on ethereum?
Theoretically those use cases are possible, but it is important to point out that you can crowdfund assassinations and create a decentralized Silk Road marketplace with Bitcoin too, even directly on the blockchain using assurance contracts. So the regulatory challenges there are fairly similar to BTC, although we do realize that the legal status of second-generation cryptoledgers is less clear; that is why we are interested in working with CODA
> Will a database of child porn be able to be stored? If there is child porn stored on the blockchain that is easily accessible, will it be legal to store the blockchain in any country?
Just like Bitcoin, the memory fees are prohibitively high to store a JPEG.
> What will be the effects of running all the clients over Tor by default? Can a 60s block time with huge blocks work over Tor?
Maybe. Wait for the testnet and try it yourself.
> [ Satoshi's arguments ]
The nice thing about having a Turing-complete scripting language is that you can integrate many interesting models for how one currency can be integrated into another:
(1) The sub-currency can be fully stored as a contract (space heavy, time heavy)
(2) The Merkle root of the sub-currency state tree can be stored in a contract, and the leaves off-chain. Each update to the tree is validated by the contract (space light, time heavy)
(3) Two blockchains, but where the Ethereum chain includes an SPV client of the other chain (space light, time light)
If one way of doing things is too inefficient, we can always fall back on other ways.
Hi Vitalik, I'll see you at the Miami conference. If you could give your thoughts on these questions, it would be much appreciated. I am very interested in this project.
NECESSITY OF A NEW BLOCKCHAIN
The whitepaper states that colored coins inherit the same limitations as the bitcoin protocol. Metacoins do not, but they cannot be truly secure without having a copy of the blockchain. How big of a problem is this? Hasn't the Electrum model worked for Bitcoin? Is there a big issue with just querying a bunch of supernodes to see if they all give the same result? Is it necessary to have absolute security with every contract, or only very high value transactions? Is it correct that metacoin clients won't need the first 14gb of the blockchain?
BITCOIN BLOCKCHAIN SIZE
Bitcoin Magazine's analysis showed that since storage capacity and internet bandwidth follow Moore's law, the blockchain size will not be a problem to store on phones. How many years from now should we predict that the blockchain size will become irrelevant? If the blockchain size were irrelevant, would that remove the need for ethereum? How much demand is there for blockchain contracts at this time, and how fast should we expect demand to accelerate?
How fast should we estimate the ethereum blockchain size will grow? What will be the effects of ethereum blockchain size growth several orders of magnitude greater than Bitcoin? Will this lead to more mining centralisation?
Is it reasonable to think the client can be sandboxed? What will be the effects of these, since the protocol is Turing complete: viruses, malware, adware, keyloggers, scam contracts, wallet stealers
What will be the effects of illegal contracts? An assassination market, for example? Will the Silk Road be able to run on ethereum? Will a database of child porn be able to be stored? If there is child porn stored on the blockchain that is easily accessible, will it be legal to store the blockchain in any country? What will be the effects of running all the clients over Tor by default? Can a 60s block time with huge blocks work over Tor?
In 2010, satoshi succesfully argued against including Namecoin into bitcoin, suggesting merge-mining instead.
"Piling every proof-of-work quorum system in the world into one dataset doesn't scale.
Bitcoin and BitDNS can be used separately. Users shouldn't have to download all of both to use one or the other. BitDNS users may not want to download everything the next several unrelated networks decide to pile in either.
The networks need to have separate fates. BitDNS users might be completely liberal about adding any large data features since relatively few domain registrars are needed, while Bitcoin users might get increasingly tyrannical about limiting the size of the chain so it's easy for lots of users and small devices." -- Satoshi
You can view the rest of his arguments in the thread here:https://bitcointalk.org/index.php?topic=1790.0;all
I have not seen any core bitcoin developers comment on ethereum. Is there somewhere besides the reddit and the thread on bitcointalk about it?
I am not very technical, so sorry if some of my questions are invalid or stated incorrectly. I also posted this on reddit.com/r/ethereum since it seems like Vitalik responds there more often.