You could try. Remember SD objects have 512 bits of address and I am not sure how the 32 bits for the biycoin # fits into that
No, farming rate is determined from the available space compared to used space so historical FR data would of necessity require the number of GETs done over time.
Also FR varies slightly between groups so any historical data will be inaccurate anyhow.
As to your scanning the coin address space, you can get a rough approximation by scanning two or three 1 million regions since coin is issued on a seemingly random distribution due to the XOR space appearing random to linear space.
Obviously the larger the space you scan the more accurate your results would be.
But of course there is one factor that can skew that approximation and that is paying for PUTs could be not random in the addresses used if the payers chose which coin to use rather than a random one. Thus my suggesting 3 regions. Also if the regions are quite different then a larger sample would be needed
For any group to know the actual number they would of necessity need to scan all the coin addresses. They only need to be able to find out if a coin exists or not, tehy don’t need to know the number. The group is only ask to issue a coin at a set address by the farming/PtD/PtP algorithms, they are only asked to transfer ownership, or asked to recycle the coin at that address.
In other words they are only asked to operate on a specific coin address. They are not asked to allocated a coin or such action.
I am sure there will be an APP created to approximate by examine a region of the coin space and update a SD holding the latest value for that region. Then every time the APP is run by someone it chooses the region SD that is the oldest and gets the count and updates the SD. Then the APP reads all the SDs and tallies up for all regions.
This APP could then spread the load across all those who wish to know and if enough do wish to know then it will always have recent data.