You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
With a database now listing header metadata by timestamp, we should figure out if the emails of the very early and very late timestamps should be included in our dataset.
For notation purposes:
The earliest legitimate email is Dec 29, 2010. Before that there are emails from 2002, 2004, and 2006.
The latest legitimate email is Nov 10, 2015. After that there are just typos (5014, etc), so I'm guessing that is the proper upper limit.
(This is maybe more of a to-do reminder for me.)
The text was updated successfully, but these errors were encountered:
With a database now listing header metadata by timestamp, we should figure out if the emails of the very early and very late timestamps should be included in our dataset.
For notation purposes:
The earliest legitimate email is Dec 29, 2010. Before that there are emails from 2002, 2004, and 2006.
The latest legitimate email is Nov 10, 2015. After that there are just typos (5014, etc), so I'm guessing that is the proper upper limit.
(This is maybe more of a to-do reminder for me.)
The text was updated successfully, but these errors were encountered: