We have had some internal discussions about the mining system. Currently the full block data is sent through to the website visitor or self-miners browser. This was useful to start with because technical users could open up a developer console and see the blocks coming in and out to make sure nothing untoward was going on. As the block size and traffic volumes increase this is becoming more and more expensive from a data transfer perspective. Last months data transfers costs were around $1000 USD and made up a significant part of our hosting bill. Now we have the block explorer in place what we would like to do is introduce a system where we keep the data server-side and just send a block number and SHA256 hash out to the miners. This will make mining quicker (less data) and reduce transfer/bandwidth. Anyone who wants to check will be able to use the blockexplorer to match up the SHA256 hash they receive with the data in the blockexplorer. So in effect miners will be hashing a hash if that makes sense?
To explain further the block data will be collected into a given block number i.e. blockchain/123, we will take this data and create a 256 bit string server-side. This string will be sent out to the miners along with a block number who will add a nonce (random number) to the end and rehash that data. Leading zero hashes will be returned to the server as normal.
So we will be sacrificing some transparency in favor of lower data transfer costs. But the system will still be fully verifiable because all the data is still publicly available using the block explorer. Would anyone have any objections to this system being put in place or any thoughts on the matter?