Edit: Initial reward/cube root(height) gives a reward of 0.0968 XMR at block 6,000,000. That seems about right, and provides 139 XMR/day for miners. Somebody else will have to math the limit on the sequence, I've been up too long.
Good formula but still too much, if XMR is at ~1000usd at block 6mm thats ~140k fresh usd that new to flow daily.
What would the "~1000usd at block 6mm" market cap be? I'm guessing it is high enough to justify/demand at least $140k/day worth of network security.
If we use inital reward/square root(height), the reward at 6MM falls to 0.00718 XMR/block, or 10.34 XMR/day. Better? $10k/day to secure a multi-billion dollar network seems far too low. Of course tx fees also have to be guestimated and included...
We have our pick of many fancy numbers between 2 and 3 for fine tuning our eternal reward with irrational or transcendental flair.
(some examples)
I'll start the voting by saying I have no preference and expect tx fees to dominate by block 6MM if the coin is a success.
1. This discussion of what the reward will be at block 6M is completely inconsequential and simply spams the thread and the funding conversation. Nekomata simply provided an example, and your examples for the exponent prove that there is no practical difference between them given the orders-of-magnitude error bars on all the rest of the data.
2. All reward profiles that decay slower than ~ height^(-2) will lead to an infinite total emission, contrary to your earlier assertion that it needs to be finite. You propose decay between ~ height^(-1/3) and ~ height^(-1/2) which clearly generate nonconverging partial sums. So you contradict yourself.
3. I disagree with changing the overall shape of the emission schedule. There are good reasons to keep an exponential schedule and changing the base of the exponential leads to either (a) accusations of greediness among early adopters for increasing the base or (b) resentment from some early adopters for "halving their coins".
4. Exception to (3) for tail emission. Now the tail, no matter how it is chosen, must not result in increasing inflation, so inflation after the tail takes effect must be lower than inflation when the tail takes effect. Thus, it is not justified to change the emission schedule before inflation reaches that upper-bound (be it 1% or any other, as would be decided).
5. Even a simple schedule like "keep current until 1%, then keep 1%" leads to some possibly unintended consequences, like a nonmonotonic block reward (decreasing until tail starts, increasing afterwards). So making these decisions is not a simple task and should not be trivialized with arbitrary choices of irrational or transcendental values for the lulz.
6. The whole point of a different tail emission is (like smooth mentioned just a few posts earlier) that miners do not depend on transaction fees. Taking this to the logical conclusion, we should assume zero transaction fees for the purpose of the tail. Furthermore, the other purpose of transactions fees if not miner reward is spam prevention. This can possibly be deflected with proof-of-work "fee" and monetary transaction fees eliminated. You keep repeating that miners should be rewarded from the fees; this is precisely what we're trying to prevent with this discussion.
All these and some of your earlier comments like the inclusion of rpietila or CZ in the group that decides how the devfund is spent (7) suggest you either are trolling or have ulterior motives.
7. There is no reason or moral ground for a representative of the large holders or a competing crypto-currency to have this kind of decision power on the devfund. If CZ joins the Monero core devs, he would be entitled to that position. Similarly, rpietila will already represent the large holders to some degree through the Workshop. Once the Workshop dispenses half of the vote cost to the devfund, the Workshop should have no decision power on those funds.
I'm calling you out.
On topic:
I am still hesitant about the postmine, but if other methods like the crowdfunding do not work and it has to be done, then I propose we simply move emission from *before* the tail starts to the present.
Say a bootstrap block with reward X to the devfund is generated "soon". Then if X represents y% of the "total" 18.6M emission, it can be covered slowly by reducing mining reward by >y% until the deficit is covered. Thus the net effect would be that the core devs make a "loan" of X from "nowhere" at 0% interest, and then repay it back by burning part of future emission. This would make the bootstrap a contained problem, since it is a localized, surgical change in the emission schedule. To the first order, any injustice will only last until the devfund would have "repaid" the "loan".
But smooth, both for the crowdfunding attempt and as justification for the "loan", that spreadsheet needs to be done and it nobody but the core team that can do it.
--> 0.02916 XMR/block6MM; 42 XMR/day
