[consensus uncertain] EIP86: initial account abstraction

[pirapira]

When merged, this will fix #231 .

[pirapira]

So far I have implemented points 1. and 2. of ethereum/EIPs#86

[chriseth]

Interpret the signature (0, 0, 0) as signed by 2^{256} does that mean that anyone can spend the balance at 2^{256}-1? My interpretation was that only the EVM returns that value for tx.origin and msg.sender, but it is not actually the sender of the transaction.

[pirapira]

I believe so at the moment. EDIT: this comment was ambiguous. I believe that anybody can spend the balance at that address.

[pirapira]

I still might need to say "the nonce of 2 ** 256 is never incremented".

[chriseth]

We should probably also disable nonce-checking for that sender address.

[pirapira]

Nonce-checking?

[pirapira]

Now I see the nonce-checking. I keep it for now.

[pirapira]

This one is the pull-request version of EIP ethereum/EIPs#208

[pirapira]

• I'm seeking clarification about the nonce increment Abstraction of transaction origin and signature.md EIPs#208 (comment)

[pirapira]

• seek clarification about the case when some code already exists at the address Abstraction of transaction origin and signature.md EIPs#208 (comment) (I just guess similar to oog)

[pirapira]

This PR will be adjusted and reopened on metropolis.