If you’re a frequent identifier you probably know how tricky it can be to fully get your notifications down to 0 and keep it there for any length of time. While I check what I can, I usually have a 3 digit number.
Logging in today I noticed that ~2/3rds (several hundred) of my notifications just disappeared overnight without me getting to them yet.
Was there some sort of update that I missed? If they removed notifications for elaborations upon coarse IDs, I suspect a similar proportion of notifications would disappear. If that happened I might celebrate.
I am not sure if there was an update but I’ve definitely noticed that if you open the notifications list and then navigate away from the page, the list is restricted back to the most recent 20 notifications or so and clears the unread marker. It can be maddening. That isn’t what happened in your case so I’m not sure the cause, but possibly related to this?
I think you can selectively disable notifications for certain things, but probably not for coarse IDs only.
Goodness, to not have agreeing ids disabled would be a landmine to find actual pings. I would get like 500 notifications a day and you have double the amount of ids as me! How do you manage?
I don’t think there’s been any direct work with notifications in a while, and I personally havne’t noticed anything off today. If you have more details that would be helpful.
I also lost approximately 200 unread notifications overnight; is there some sort of “timeout” if the notifications are left unread for too long? I definitely did not open the menu to read any, this happened with no input as the OP describes.
sorry to say, but while I already have agreeing ID notifications disabled, I still definitely do not have time to address all of my notifications per day at this rate! a lot of pings addressed to me take more thought to answer, a perusal of my reference library, and so on… if I were getting 30-80/day it would be absolutely impossible to catch up, and even at my more like 20-50/day I have not been making headway.
so in these circumstances losing a large chunk of notifications is kind of disastrous.
I’ve already been using this to whittle it down and choose which ones I reply to first and most quickly – in fact this tool is how I know that it’s my oldest notifications that were lost. still, I haven’t been able to get through all of them even with this tool, which makes iNaturalist notifications bearable instead of simply impossible to use. I do highly recommend it!
Definitely sounds similar to my experience. I had pondered the idea that maybe there was a timer and I just took too long to look, but if there was, the day that just timed out would have been some sort personal record for most notifications, but I don’t remember any corresponding leaps of the same size. 200 notifications a day is not unheard of if I’ve IDed unknowns this week, but usually that would be the most I’d get a day, and only if I’ve just been into the unknowns section. But this was much more than the “usual most” I expect to recieve.
And unfortunately agreeing IDs are already turned off. To lower my notifications further I would need an option to get rid of “elaborating, complimentary IDs” or something of this nature.
I also use iNat Next when on my phone, but I still use the notifications tool when on desktop. between them my rate of reply has definitely improved… yet still not enough.
I wonder if there is some sort of soft threshold? I have to confess, this is not the first time this has happened to me. I lost about 100 or so notifications a few months ago – but at that time a prolific user I knew deleted their account, so I assumed I had just lost all notifications associated with their observations or comments. but the fact that it’s happened again, if perhaps for a different reason, did make me wonder. I agree with your thought that it’s too much for a single day and that there must be something else going on. I don’t really have any additional insight unfortunately.
Sorry, I need to correct this. Old notifications are supposed to be kept in the database for three months. However, there was a bug that didn’t run that job which was just fixed, and it may be the cause of @benkendrick’s issue in the original post here.