This is one of those questions that feels so obvious to ask (as a total beginner) that I’m sure the answer is similarly obvious and I’m just not looking in the right places to find it:
Why don’t we have some sort of “global” or “universal” dichotomous key for the purpose of identifying living things down to… maybe just the family level if individual genera/species would be too complex?
It seems like, especially with the aid of modern technology and databases like inat, it would be pretty feasible to design a tool that consolidates large amounts of taxonomic data (specifically what traits differentiate each phylum from one another, each class, each order, etc.), and then turns it into a mostly text-based interface that anyone could use to at least narrow down what kind of organism they’re looking at. I think this is kind of what our current CV model does (in a very very roundabout way), but I don’t see why it couldn’t be turned into a more key-like format.
I understand that nature is weird and full of exceptions to rules, as well as situations where two species can only be differentiated under the microscope or with a population range map, but at least for those organisms that can be easily identified, why don’t we have a tool like this? And if we do… where can I find it?
One thing you find when writing identification keys is that the larger the area covered, the more difficult the key will be to use. Why? Because you have more species that are similar to each other, so you have to use more inconspicuous traits to distinguish them.
Example: In North America there are three recognized species of yellow-breasted meadowlarks (genus Sturnella). In the east, there is only one species, the Eastern Meadowlark, https://www.inaturalist.org/taxa/9528-Sturnella-magna. Easy to key out. But if you’re making a key to all North America, you have to distinguish the Eastern Meadowlark from similar Western (https://www.inaturalist.org/taxa/9535-Sturnella-neglecta) and Chihuahuan Meadowlarks (https://www.inaturalist.org/taxa/1456513-Sturnella-lilianae) and the differences are subtle. If you’re trying to write a key for the whole world it gets more complicated. You may find yourself having to key the yellow-breasted meadowlarks (Sturnella) from the Yellow-throated Longclaw (https://www.inaturalist.org/taxa/14104-Macronyx-croceus), unless you’ve already keyed Sturnella away from Longclaws and their relatives earlier, using much more basic characteristics that may be hard to understand or to see.
So, although a global key seems appealing, for most species it’s more practical to use a key focused on as small an area as is likely to include all the possibilities.
The larger the area considered, the higher species number and the more subtle differences. In other words, a dichotomous key with a very high number of species would be very hard to be used.