Ocean West Posted July 13, 2020 Posted July 13, 2020 Haven't paid much attention to statistics that the ACP provides but I am curious about how the background tasks of auto Prune play a role in the percentages. And wonder if there is any option to delineate negative values based on auto prune or end user action?
bfarber Posted July 14, 2020 Posted July 14, 2020 If you prune data, then it cannot be incorporated into statistics. Hopefully that answers your question. For instance, if you prune old follows (looking at the screenshot there), then the statistic values will adjust as the data is pruned. We don't store historical data separate from the raw data that is used to generate the stats.
Ocean West Posted July 14, 2020 Author Posted July 14, 2020 So if i have delete follows from 1000 days ago every day is a rolling set of new potential people who now qualify for Pruning. really anything that is time specific as time moves forward will scope a new set of users that qualify on actions. Perhaps a s feature - future forward looking stats that today we prune 50 tomorrow 23 day after 32 next month 1224 etc.
bfarber Posted July 15, 2020 Posted July 15, 2020 Pruning follows only impacts stuff that hasn't been updated in a while, so it would be impossible to calculate that. What if a topic gets replied to tomorrow (after not receiving a reply for 1000 days)?
Recommended Posts