2023-04-20 UTC
# 
bkil We could actually compute individual, per-user stats to compute weighted peak times (after work or breaks?) and downtimes (sleep patterns?) to arrive at approximate time zones for everybody and then aggregate stats based on that derived table as well.