I wimp out and downvote wrong annotations. Not helpful I know.
There is an open feature request to make annotations votable like Community Taxon.
I wouldn’t say that it isn’t helpful. For most people, that’s a reasonable approach. Because I’m using the observation data, I need a way of overriding mistakes that I can’t correct directly. Observation fields allow me to do that. It isn’t really helpful to anyone else unless they recognize the correspondence between the incorrect annotations and the “corrections” I’ve entered via the observation fields.
And now that I have a way to enter additional fields in an efficient manner, I’m exploring adding additional information which isn’t covered by existing annotations (eg. seasonal morphs).
And now it looks like the observation fields may give me a way of ignoring problem observations.
2 Likes
Common sense would just suggest not to ID them.
Yes, we can spend some time suggesting users to put some care in making better observations, maybe in revising the photo(s) they are going to upload. But, it is worth doing it? Maybe with some users, but with other it is just a waste of time.
In conclusion, a wrongly identified bad observation, maybe even raised to RG, is much worse than an unidentified one.
Fair enough, but if you are methodical, and you ID all the observations that fall into a certain set of geographic/taxonomic categories (for me, it’s all buttterflies reported in the province of Ontario), you might not want to have a bunch of observations sitting in your “queue” that you keep checking because you don’t remember why you didn’t identify them earlier. Say an observer submitted 10 photos of the same species/individual photographed at a particular date/location. The first time you see the series, you might decide you’ll ID the first one and skip the other 9. Three months later, you might see the series again and think “these look familiar, but I can’t remember if I already ID’d one of them”, so you ID another. Then it happens again a couple of months later…
Another potential problem is that an observation could pop into your queue of unreviewed observations long after you’ve passed that point (maybe it was languishing at a higher taxonomic level for a long time before somebody ID’d it to a level where it becomes part of your queue). If your queue of unreviewed observations is clogged up with dozens of garbage observations, you aren’t likely to notice a new identifiable observation that appears in the middle of the mess. And if you’re paranoid about missing new/old observations like this, you’re more likely to waste time revisiting the garbage observations, looking for anything that you might be overlooking.
If I have a mechanism by which I can “exclude” garbage observations from my view, then I don’t waste time re-visiting them at a later date, and there’s less chance that I will overlook a new observation that (belatedly) pops up in my queue.
As I said, I’m using “reviewed” to indicate that I consider an observation to be valid, so I can’t click “reviewed” to skip a garbage observation. I suppose an alternative approach is that I could treat my identification as the indicator that an observation is “valid”, but that would require that I “agree” with a lot of observations that are already RG. That takes more time than clicking “reviewed”, which I can apply to an entire page of thumbnails at a time.
3 Likes
This is my approach. I don’t find it too time consuming since I look at each observation individually in Identify (and agreeing is the same amount of time as reviewing). I used to use the thumbnail view in Identify but found I was missing too many errors on tricky observations with low quality photos.
1 Like
I could see that happening with some taxa, but with butterflies, the bulk of the observations are of a handful of common species (a ridiculous percent of the observations are of Monarchs). When reviewing observations that are already RG, I can usually look at a page of thumbnails, click on a couple of the ones that require a closer look, and then just click “set all as reviewed” and move on to the next page. In peak season, I have to review hundreds of observations per day just to keep my head above water. Maybe the bulk metadata tool could help with this, but I haven’t had much luck with using it in bulk mode (so far). It’s great for adding annotations and obs fields to individual observations however.
Filtering based on the presence of an innocuous obs field appears to work fine so far, but clearly, we wouldn’t want a bunch of people adopting this method and cluttering up observations with a bunch of random fields. But if that happens, I guess it would demonstrate that there is a clear need for some form of “ignore” checkbox, and maybe it would be added.
1 Like
The reviewed button exists so you can choose never to see the observation again. You’re having this problem because of the way you’re using the reviewed button. To anyone starting out, I’d say, “Don’t use the ‘reviewed’ button as an ‘agree’ button!” But you are deep into this. I recommend choosing some field (even if you have to make it up!) that you mark these with and then filtering out these observations when you’re IDing. Or add another ID to ones you agree with! Takes a little more time, yes, but it adds value to the observation, too, assuming you know what you’re doing. I find it doesn’t take a lot more time if I open a page of photos in “Identify” and just quickly click on the “agree” button on each thumbnail. If you’re opening them anyway, it’s an even smaller percentage of the time you’re spending on the observation.
3 Likes