--- Log opened Sat Feb 29 00:00:09 2020
07:41 < bsm117532> One consequence of the above "average work of parents" rule is that a cohort has an effective work that is the work of the longest single subchain.
07:42 < bsm117532> An RMS average (or higher p power mean) more strongly weights that sub-chain. (I'm still unsure this is desirable)
07:44 < bsm117532> Another consequence is that a multi-bead cohort always has an aggregate work that is <= that longest single sub-chain.
07:45 < bsm117532> That's unfortunate since it seems miners aren't incentivized to include stray tips, since it will lower the chain's overall work.
07:46 < bsm117532> One of the power means is p=\infty, which is the simple max(). So, max is in the category of "power means" and can be used instead. This just selects the longest single sub-chain, completely ignoring work outside the longest sub-chain.
07:50 < bsm117532> Note all the above is only for deciding double-spend resolution, not rewards. All miners would receive a reward proportional to the work, regardless of the position of their bead in the DAG (up to a latency factor in which their bead would most likely be a bitcoin orphan).
07:51 -!- belcher [~belcher@unaffiliated/belcher] has joined #braidpool
16:34 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 255 seconds]
16:37 -!- belcher [~belcher@unaffiliated/belcher] has joined #braidpool
18:29 -!- belcher [~belcher@unaffiliated/belcher] has quit [Quit: Leaving]
--- Log closed Sun Mar 01 00:00:11 2020