changing the emission curve to favour the early adopters directly is taboo. It should not even be on the table.
I agree. But we are still left with the question of what favors the early adopters (presumably this means over later adopters?) and what does not.
Example:
I mined in block 10 hold on to my coins, never buy or sell, and never mine again.
How does shifting around the rewards in various ways (unspecified) between blocks 11 and 1 million
but without changing the total number of such coins mined affect me one way or the other? I suppose there are extreme cases where it likely does. For example, moving all the rewards to block 11 and then having no rewards. In that case, it is likely no mining occurs and the coin dies, so it does affect me.
Leaving out those obviously harmful permutations, how does this affect me as an early adopter?
I have argued above that the bolded phrase is meaningless. Please read my post again.
EDIT (to prevent double post): I proved maybe ~50 pages earlier that Monero's block reward is actually a simple exponential decay once you solve the recurrence relation. This means that the emission curve is really only defined by three parameters: constant term (premine, 0), reward of first block (17.something but can be normalized to 1 for this discussion) and base of exponential (reward "decay factor", which is 1 - 2^-something * 10^-something, a value very close to 1). I claim there are good reasons for the exponential form and should not be changed. Now let's consider our options:
1. A post-mine changes the constant factor and a posteriori (i.e. for later adopters) it
is the same as a premine, which is frowned upon. A 1% premine seems to be acceptable, but one of Monero's main selling points was ZERO premine. Even 0.1% is much larger than 0. As I noted before, I would accept a 1% post-mine but only if other methods of fundraising have failed.
2. Changing the normalization factor is equivalent to a redenomination. This would have been acceptable in April (second option in my previous post) but I do not believe it is the case now. I think doing this will kill the coin with p > 0.5. Increasing the normalization factor is the second option in the previous post, decreasing the normalization factor is a posteriori equivalent to a fastmine.
3. increasing the decay factor is a posteriori equivalent to a fastmine. See (2) above. Increasing the decay factor is the same situation as decreasing the normalization factor.
So to sum up, any of [increase|decrease] the [normalization|decay] factor, as well as increasing the constant factor, clearly and directly affects one of [early|late] adopters to the advantage of the other group. Fucked if you do, fucked if you don't. And in this situation, the only winning move is to not play.