I would like to contribute the following post to
https://bitcointalk.org/index.php?topic=114664.40 :
As you can hopefully see, I have a significant amount of knowledge about the different coins.
Thanks in advance,
My post:
Bitcoin is brilliant except for to senseless calculations everybody is performing to mine a block. It's great to see that this single oddity has also been resolved using coin age in PPCoin. However, it's clear that there's still the issue of missing incentive to process transactions when minting a new block.
Many possible solutions have been suggested already, hereby another one:
Let's say that for example on average 2880 coin days per day (24 hour) are destroyed in transactions, and let's also say that it is desired that every 5 minutes a new block is minted for fast transaction processing, on average every block should then process transactions worth 10 coin days.
The system could then mandate that only blocks that process transactions worth at least 10 coin days are valid blocks. Just like Bitcoin's difficulty this amount could be adjusted every x blocks: if it turns out after x blocks that there are not enough coin days destroyed in transactions to reach a 5 minute average, the mandatory amount is lowered and vice versa. There is no need to pay any incentives to minters, to earn your 1% interest you must process transactions.
There is also no need to process more than (in this example) 10 coin days, because this will not increase the amount of interest you will receive.
The amount of coin age contributed to the blockchain by a block is the sum of destroyed coin age of the processed transactions and the destroyed coin age of the minted coins (interest receiving coins). This will ensure that every minter will still be able to collect its 1% in due course and that older transactions will have a higher likelyhood of being processed because of the building up of coin age the longer they are not processed.
However, there is one problem with this scenario (or perhaps more that I'm not aware of). A minter could decide to process more than the required amount of transaction coin days, to increase his chances of ending up in the blockchain or to perform a little bit of sabotage by submitting a block with lots of transactions forcing a reorganization of the blockchain.
This problem can be a countered with the following measure:
The amount of coin age contributed to the blockchain by a block is the minimum required amount of processed transactions (in my example 10 coin days), regardless of the actual amount of processed transactions + the destroyed coin age of the minted coins (interest receiving coins) + fixed amount per block (this is not a reward to anyone, but only a contribution to the total coin age of the blockchain).
Let's look at an example to see what this will achieve:
Let's say we have 30 coin days worth of transactions to process, and the minimum required amount is 10 coin days. Let's say the fixed amount per block is 5 coin days, and all the actors combined per fork have the same amount of coin days to mint, for example 7.
In fork A they only process the minimum required amount of coin day transactions per block (10 coin days), in fork B they process more (15 coin days).
The total amount of coin days contributed to the blockchain in fork A is: 10 + 5 + 10 + 5 + 10 + 5 + 7 = 52 coin days.
The total amount of coin days contributed to the blockchain in fork B is: 10 + 5 + 10 + 5 + 7 = 37 coin days.
One can even do without the fixed amount per block (5 coin days in my example), but including the fixed amount will indicate sooner which fork will win.
It is clear that in this scenario, to maximize his chances of ending up in the blockchain, it is in the minters best interest to stay as close as possible to the minimum required amount of coin day transactions per block.
One exception is necessary: when a single transaction exceeds the minimum required amount of coin day transactions per block, the full amount of coin days of this single transaction should contribute to the blockchain, to prevent this transaction from never being processed (it will now be first in line to be processed).
