to mathematicians, this notion of the tail emissions, which really just means interpolation of the stair-step decline of rewards, as an attack, is ridiculous. if the graph is plotted out, and the number of sats at each block is the same in both cases, there has been no change in supply. it does, however, mean that likely the subsidy will continue for a few years after the halving pattern would have dropped to zero. but that extra would literally be the smoothed out segment of the last reward cycle moved to the next. it would only extend to the next cycle after what would have been the last cycle, and only part way into it. what is more important is the complexity of implemenatation tail emissions curves are an exponential decay cycle which means transcendental math. the risk of forks being caused by this is substantial, the algorithm is complex and there is not many common fixed point fractional exponentiation implementations in existence. not sure why, but without a deterministic formula for it, i say better to stick to halving.