The cutoff time was c.notMisleadingUILaunchTime, not the TombstoneLaunchTime.
Other changes to docstring just to make it easier to follow. Please check my understanding, that before c.notMisleadingUILaunchTime, you didn't assess helpfulness of notes that didn't assert the tweet was misleading, but after that you did.
Added a line to the logging to show the quantity for the last group of ratings you're keeping.
The cutoff time was c.notMisleadingUILaunchTime, not the TombstoneLaunchTime.
Other changes to docstring just to make it easier to follow. Please check my understanding, that before c.notMisleadingUILaunchTime, you didn't assess helpfulness of notes that didn't assert the tweet was misleading, but after that you did.
Added a line to the logging to show the quantity for the last group of ratings you're keeping.