Components of size should be part_of rather than Is_a children

[cooperl09]

The components of size (PATO:0000117); the terms for 1-D extent, 2-D extent and 3-D extent should all be should be part_of size, rather than Is_a children.
Currently in PATO, length, width and diameter are all direct children of size.
Thus leads to weird inferences, such as stem diameter (TO:0020083) is a stem size (TO:0000892), in the Planteome.
See: Planteome/plant-trait-ontology#447

[matentzn]

Hey @cooperl09

Not quite sure I follow: in my (albeit very limited) world view diameter is a kind of size; or are you referring to the fact that an increased diameter does not automatically imply an increased size?

[cooperl09]

IMHO, an increased diameter is only one aspect of increased size. Height, length, thickness etc are all aspects (part_of) of size as well.
This assertion leads weird looking inferences, such as stem diameter (TO:0020083) is a stem size (TO:0000892), in the Planteome.

[matentzn]

Ontologically speaking, you are probably right. Although increased diameter is only one aspect of increased size is probably not quite true as you can have an increased diameter while having a decreased size.. But your general point holds; I assume it comes from the intuition the qualities here sound a bit like measurements; it makes more sense to say: a measurement of diameter is a measurement of size. But since a PATO class is NOT a measurement, probably you are right. lets see what the vanguard says. @dosumis @balhoff @cmungall

[cooperl09]

It also seems odd that diameter (PATO:0001334) is a child of length (PATO:0000122).
Thus leads to weird inferences, such as stem diameter (TO:0020083) is a stem length (TO:0000576), in the Planteome.

[cmungall]

much as I hate proliferation of similar terms there is an argument for a 'size attribute' grouping class and then being stricter about size as Laurel suggests

[matentzn]

I am not comfortable with these issues and defer them to my boss :P @dosumis