it seems the cursoring as discussed in https://dev.twitter.com/docs/working-with-timelines suffers from an intricate problem.
Consider consuming a timeline through statuses/user_timeline?count=200&exclude_replies=true
if the consumed timeline happens to contain 200 or more subsequent replies, the previous API call will not produce any results (i.e., ). Thus, there is no new max_id value that can be chosen for correct cursoring and no further progress through the timeline can be made.
Unfortunately, this is more frequent than one would assume. For example, customer relations accounts frequently contain very long sequences of only replies.
Furthermore, this situation is indistinguishable from reaching the end of a user’s timeline which also returns .
Please advise how to implement cursoring in a correct way.