if it is done this way, it would increase efficiency by adding computing power to the most recent block and offering higher reward if you truly want to sync all the way back down to the... Thinking about it older blocks take more work and should have higher reward for storing but newer chains lower reward but more consistent and faster
@Arachnid Actually a better approach would be just have a DHT(simple table of block id hashes) of blocks/epochs/dynasties then take the top order chain and start processing in reverse order making more recent transactions priority... Building everything from first to last seems inefficient as the first few blocks will be accessed the least. Is my thinking in line with your philosophies, i'll be happy to work on an EIP if this is feasiable
In either case, shouldn't reverse order building seem like an ideal approach as more recent blocks will be higher priority
@jamesdigid How would you process blocks 'backwards'? Transactions aren't reversible.
I suppose we could add the actual ETH blocknumber there but a few clients make it possible to enable EIPs starting at your own custom block numbers. So I supoo it should be clear that it's the ETH fork block
I recommend either adding the fork block number in place of TBA, or removing the the FORK_BLKNUM line all together. Having it there but TBA implies that it isn't actually a live change. (I strongly preference the first option of correctly setting the block number).
ethereum/EIPs#833 fixes this issue.
@nicksavers do you think it's a good idea to include the "previous fork" in fork EIPs? For example, the DAO Fork EIP mentions "Previous version: Homestead".
The bytes are sent to a different function after the call goes through
This is a composition of sending and performing some other function
ethereum/eips#677 also includes provisions to ensure that a contract which receives tokens should be prepared to accept them
Does anyone see value in this?
I do not see value in this because the approve / transferFrom does this in an easier way
Does anyone see value in function composition being built into 721?
Sorry ignore ^^^
Hi, I wrote a draft spec for typed errors using REVERT. Would love to get feedback: ethereum/EIPs#838
is there any repo that implement the DAICO model already?
Hey everyone! I'm working on a universal search interface standard ERC that would allow exposing a public interface for (yet-to-be-created) Ethereum search engines. Could you please quickly check it out: ethereum/EIPs#857 ? Is it a dumb idea, or is there something in it?