That sounds like a good reason, but I still have two problems with it:
The number of observations like this is pretty small and I can’t see them having a huge impact on GBIF. There are more than 27 million RG observations which I assume are all sent to the GBIF, but if you look at Casual observations that are wild and have photos, there are fewer than 150 thousand of them. And this is still an overestimate because it includes observations that are casual because they lack locations or dates.
Surely these observations can be marked RG without them being sent to the GBIF. If GBIF doesn’t want them that’s all well and good, but marking them as casual on iNat just makes them incredibly difficult to find, and anyone who doesn’t have some experience with iNat is probably unlikely to find them and use them. To find those observations that I linked to above, I had to change several filters and manually change the URL, and I still couldn’t exclude observations that are marked casual because they lack dates or locations. It just makes these observations extremely difficult to find and use when they’re marked as casual.
Not all of such observations are marked yet, e.g. tons of Poacea that can’t be ided even to genus. Ideally the will be marked. People just usually are not pleased to mark something casual.
@fffffffff this is exactly why I think there needs to be a keyboard shortcut in the Identify pages, so that these masses of unidentifiable observations can get cleared out more easily
I regularly use the “No” box and find it a useful feature for providing some quality control when I curate taxa. I primarily use it for observations whose photographs are too poor to ever be reasonably identified. I have a fairly low bar for what I consider an acceptable photo, but some users will post literally any blurry photo their camera captured, often making preposterous IDs from them. These just add clutter when scrolling through the observations of that taxon. I know the counterargument is, “ah, but what if an expert can identify it from a blurry photo”, but, again, I set a pretty low bar. I think as long as this feature is used respectfully, it is advantageous.
Bugguide provides a similar tool for its curators, which allows them to “Frass” (i.e. delete) low quality observations, while keeping the data associated with them. In this way, the functionality as a photo guide isn’t compromised by a bunch of blurry, barely recognizable images.
I fairly often find observations with 5 or 6 species-level IDs that are still Needs ID, where it’s obvious the original observer clicked the box without knowing what it was for. It just takes one click on Yes to make them Research Grade and remove them from the Needs ID pool.
What I would prefer to see, rather than change the number of votes needed, is that the observer should be notified in such cases - which seems to be on the way as part of the Notifications revamp.
The “can it still be confirmed or improved” will move an observation to RG if that observation is identified at genus level or lower. If it’s family or higher, it goes to casual instead.
That used to be the case but it was changed a while ago. It won’t have any effect until there are two IDs (since there technically is no community taxon until there are 2 IDs).
