Closed Bug 1482965 Opened 8 years ago Closed 8 years ago

Decide on mitigation strategy for active_ticks in longitudinal

Categories

(Data Platform and Tools Graveyard :: Datasets: Longitudinal, enhancement)

enhancement
Not set
normal

Tracking

(Not tracked)

RESOLVED WONTFIX

People

(Reporter: bugzilla, Unassigned)

References

Details

Per Bug 1482466, active_ticks is computed incorrectly for subsession_count >= 3, which is ~8% of pings. There are a number options for mitigation strategies for longitudinal, especially since we look at each client's pings in chronological order: - The quickest fix to get us to a more-correct number is to use the scalar value whenever possible, which leaves us with only ~2-4% of pings affected on recent dates - We could, theoretically, try to correct the active_ticks values as we walk through each clients' pings. This could, however, lead to strange and undesirable effects, however, in cases where we have shared profiles, etc. - At the very least perhaps we should mark affected pings Some inputs we need before making a decision re: how much work we should put into this, and how: - How much does this issue affect data science workflows and decisions? - The longevity of the Longitudinal dataset in its current iteration is in question -- how much longer do we anticipate supporting the job? - How much work would it be to walk session pings for an attempted active_ticks correction? - Would it make more sense to leave the current active_ticks column as-is and create a new column for corrected values?
ni? saptarshi re: How much does this issue affect data science workflows and analysis?
Flags: needinfo?(sguha)
Personally, i use active ticks in main summary which i believe is not(much less?) affected ? As for longitudinal i do not use it but others do. I'll add others to this bug.
Flags: needinfo?(sguha)
Ditto sguha, I do not use longitudinal and use main_summary for most analysis.
Ditto, I never use longitudinal anymore. Main summary is my main data source for analysis
I have dashboards that use longitudinal not but not the active_ticks column; nulling the column would work for me.
Based on what folks are saying about usage, I'm declaring this a wontfix
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → WONTFIX
Product: Data Platform and Tools → Data Platform and Tools Graveyard
You need to log in before you can comment on or make changes to this bug.