Is there a way of differentiating between folders and datasets?

[CompuGlobalHyperMegaNet]

I was just creating a CIFS share of a dataset in which I store family pictures but I've noticed that when I select the Path, there are two locations in the drop down menu that do not appear as a dataset in the Storage tab. I then opened a shell via the WebGUI to double check the paths. These are the paths of the datasets and suspected folders.

Datasets:

Code:

/tank/Pictures_and_videos   (note the lower case v)
/tank/dozer_backup/dusan

Folders:

Code:

/tank/Pictures_and_Videos	(note the upper case V)
/tank/dusan

I should say that I passed the "I thought my data was gone but I just mounted a dataset over the top of it" right of passage years ago, so my theory is that the "folders" if they are folders, were created at some point when I mounted then into a Jail and used the Create directory option... that's just a theory though. There's part of me that thinks they're almost certainly folders and I could just delete them but as my "Pictures_and_Videos" dataset contains family pictures and videos, and as I'd rather avoid having to restore from and outdated backup (and reimport and edit the pictures from my DSLR / camcorder) I really don't want to take any risks without having as much information as I can get.

So here are my questions-

1. Is there any way of definitively ruling out the possibility that "dusan" and "Pictures_and_videos" are datasets?
2. Is there a way of differentiating between datasets and folders, be it on Windows, Linux or through the FreeNAS WebGUI or shell?
3. Should I just snapshot everything, delete the selected folders and check if the shit hits the fan?

 

[MrToddsFriends]

What I would do:

Shell: zfs list shows all datasets, including those on freenas-boot.

Classic GUI: Storage -> View Volumes shows pools and datasets, excluding freenas-boot.

Does this leave any ambiguities?

 

[CompuGlobalHyperMegaNet]

Note: for the sake of clarity, from this point on I'll be using the CMD tag for commands and when I'm referring to a dataset by name. When referring to a folder by name, I'll use plain text.

Does this leave any ambiguities?

Rationally speaking, I would have to say no. The mystery folders in question don't appear when I use zfs list or in the storage menu. They only appear via a folder browser or when I use ls in a shell (as they should if they're folders).

But there is one thing that's making me doubt my conclusions. When I navigate to the tank dataset share, sometimes clicking on the folder in Windows Explorer in order to view the contents of the Pictures_and_videos dataset works. I can see the files and folders contained within the Pictures_and_videos dataset but there's nothing in the "Pictures_and_Videos" folder but then it can be the other way round; I can see the dataset files in the folder but not the dataset itself. Or I might be able to see the files in both at the same time or neither of them. The one behaviour that remains consistent regardless of whether I can see the contents of the dataset/folder is that when I click navigate to the dataset via Explorer and then click back, it's always the folder (and not the dataset) that is highlighted as if I'd clicked on it.