Using Computer Vision suggestions even when we know what it is

I always use the Computer Vision suggestion if available in the drop down, even if I know what the subject is - it’s just a lot faster than typing it out usually. Is this a problem? Is this throwing off user data in some way?


No, not an issue. Lots of people do this.


Computer vision is fairly computationally intensive so it will increase the server load & power consumption. Some people might assume that you were relying on the CV for the identification (though they shouldn’t because as @reuvenm says lots of people use CV as a shortcut).

Someone pointed out that you don’t need to type in the full species names, the first few letters of each word will usually bring up the species so “eu rob” or “eri rub” will both bring up European Robin (Erithacus rubecula). I think that is slightly quicker that using the CV suggestion so I have started doing that instead.


From what i can tell, if you click on the species field while in the observation, it runs the algorithm either way, as you start to type. So this may not matter too much. The site admins could speak more.


This is true. As soon as you click the species name input the CV suggestions API will be called, and the results will be cached until you visit a new observation. So don’t worry about the backend performance implications.

This is a common use case and I believe a lot of people, including myself, use the CV results this way: as a “no-type” autocomplete. The identifications will be marked as having come from the CV suggestions, so if you don’t want that little ID indicator you’d have to type the name yourself and choose a result from the search. But we know this is a common use case, and would consider it when doing any analysis of CV-created IDs.


This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.