My question about timezones was asked previously in the forum but apparently not answered: How do I change the timezone associated with a batch of observations?
When the new settings dialog was announced, I changed my default timezone to the correct timezone. Prior to that, I didn’t realize I had control over the default timezone. Consequently, I have 1000s of observations with an observed date having an incorrect timezone.
It turns out the incorrect timezone I had been using all along is offset from the correct timezone by one, that is, the observed dates are off by one hour. Maybe that’s not a big deal, I don’t know.
I don’t think you can alter the timezone independently from the actual time on multiple observations… any bulk edit is likely to make the time the same on all. For a one hour difference I would just let it slide… if you felt it was worth the effort, you could bulk add a description that describes why the time is out by one hour… or perhaps a tag or field alluding to the “time is out by one hour” status. The later could be explained in a journal post, and then if anyone asks what the tag or field is about, you just reply with the link to the journal post (rather than explaining repeatedly).
I tried it before I read your suggestion and it seems to work (sort of). I changed the timezone on 81 observations without disturbing the rest of the observed date. However, the operation terminated abnormally with 504 Gateway Timeout. I guess there were too many observations in the batch.
I have about 4,800 observations with an incorrect timezone. Given the above result, I don’t see any reasonable way to change the timezone on all of them.
Yeah, for one hour, I wouldn’t worry too much. Worst case scenario an observation around midnight ends up being off by a day. I don’t think this would be a serious impact on most research uses for any data.
check the results though. I recall something about a bulk edit, or perhaps it was a csv download, generating an error even though it completes satisfactorily.
Yes, good point. Sometimes I get that error too and then check back later and everything has been changed. (The better solution though it not to try such big batches at once.)
Yeah, you’re probably right, but it’s annoying. I wish I would have caught this earlier. The onboarding process should have forced me to set a default timezone. I had no idea that contributed to the observed date. It should be taking the observed date from photo metadata where possible, right?
They won’t be corrupted… the time out error is more about how long your browser waits for the response that it ran ok, but the processing at the other end is not dependent on your browser waiting to find out that it completed ok… I was more suggesting that if you do check, that you’d find that they ran ok despite the error… :)