are you using only the green buttons? are you sure the final navigation prior to you noticing a problem isn’t via the browser’s back / forward navigation?
or if you are using only the green buttons, do you see the problem only when clicking immediately after making some sort of update to the original observation (ex. add an identification)? (that might be the sort of thing described in a previous bug report: https://forum.inaturalist.org/t/problem-with-previous-record-functionality/3938 .)
i can occasionally reproduce the problem this second way if i do the following:
start on an observation with the green buttons enabled and the browser window throttle to (slow) 3G speed.
make a new identification on the observation
a split second after the identification is submitted, but before the original observation is updated to reflect the new identification, click one of the green buttons
In my case I am only using the green buttons. Since I usually only have the problem while uploading, as in going through a batch I just uploaded while more load in another tab, it might be the same mechanism as that previous bug report.
I have started using the identification tab instead to go through and no issues there.
just for reference, the peregrine falcon observation was updated last at 2024-08-05 11:42:09.853-05:00 via an identification by you. that’s roughly an hour before you published the original post on this thread at 2024-08-05 12:42 -05:00.
an update to the observation like this followed almost immediately by a clicking of the green arrow button is the only way that i can see that it’s possible for the green arrow buttons to be involved in triggering the symptoms of the problem you’re describing.
this – and the weirdness if you go back and forth between observations using the browser back and forward buttons – would be basically the same kind of problem related to some sort of race condition. it would be the same thing described back in the day in https://forum.inaturalist.org/t/problem-with-previous-record-functionality/3938.
for what it’s worth, race condition problems often are relatively hard to correct because they can be hard to trigger and could require a complete reworking of how a particular screen works. so would suspect it’s not actually worth trying to troubleshoot / fix this particular problem, since the effects are so minor. (if it were up to me – which it’s not – i would just merge this with the previous thread and just close them both, noting the cost / benefit calculation. often it’s best to wait until you’re planning to make another major change or redesign of the screen to address the underlying problem.)
well, if you think that’s the case, then you should document some steps to reproduce the problem this way. without some steps to reliably reproduce, i highly doubt there’s anything that anyone can reasonably troubleshoot going down this path.
I can replicate in this situation. I agree it’s not really something worth fixing, if that’s even possible, so I’m going to close this topic. Please use the Identify page if you want to go through observations relatively quickly and add content to them.