1) In the device & browser, there is a "Country" field -- am I correct in understanding that it does NOT effect how time is reported on the watch nor how data is time-stamped on the tracker or Fitbit's cloud?
2) I'm trying to confirm I understand the overall data flow and how values are stored and transferred from one source to another… Is the following correct?
i) On Phone App or Web Browser: The time-zone is set either manually or automatically. Let’s assume it is set to UTC -5 (New York).
ii) On Device: Sensor data is tracked according to whatever "local time" it has set for itself. So if the tracker gets first set up and synced with the phone in New York at 5:57pm on Tuesday (local time), it will start recording data as 5:57pm Tuesday and continue recording onward while it has charge. Once the memory is reached, the earliest data is written over and the 7 day sliding window just moves forward in time. Daily aggregates are calculated are reset whenever the local time reaches 12am (as based off of what @Rich_Laue said in this post).
iii) When the tracker syncs with a phone, it uses the set “UTC offset” to convert the local time into UTC standard time and that is sent to the Fitbit cloud.
iv) The Fitbit cloud stores all data as UTC time, but then reconverts the data back into “local mode” (New York UTC-4 currently) across the various applications:
- For the phone visualizations (local time - pretty sure of this)
- For the web browser dashboard (local time - pretty sure of this)
- For the web api pulls (local time - not sure, is this correct? API pulls are all in the given user’s local time?)
If the person moves to Europe and the “local-time” is changed to Amsterdam UTC+2 (as in step i):
- The change in local time is only transferred to the phone upon sync.
- Prior to the change in local time being processed on the tracker, all the existing data on the tracker is first transferred to the cloud according to the PREVIOUSLY set local-time (New York).
- The tracker than registers the new local-time and starts recording values according to that new baseline.
- Is that correct?
3) If all the above is correct...and one uses the API to pull the user's sleep data from both NY and Amsterdam, that all the activity/sleep/HR etc which are associated with their time in NY would subsequently be pushed FORWARD 6 hours? (e.g. if they were generally sleeping between 11-6am NY UTC -4 time, that because their local time is now Amsterdam UTC+2, that all that data is subsequently now represented is now reported via the API as being between 5am and 12pm?)
