So I think what this saying, rather indirectly (from the other direction), if I'm understanding correctly, is that the relationships that can be inferred from looking at a taxonomy are ambiguous, because a taxonomy includes 2 kinds of relationships, but encodes them in the same way (conflates them together as if they were both hyponyms--er, well, this is saying that the are both kinds of hyponyms):

• "An X is a Y" (simple hyponymy)
• "An X is a kind/type of Y".

Actually, I may have read it wrong / misunderstood it... While it's not ruling out that simple hyponymy may sometimes be used in a taxonomy, it is be saying that the "second relation" is "more specifically under the concept of taxonomy" ... which is not really clear, but seems to mean that it is more appropriate / better for use as a criterion in a taxonomy.

Okay, so define "simple hyponymy" and name the other kind of hyponymy that is referenced here.

Good explanation about semantic fields.

I assume the same or an even stronger statement can be made about semantic classes (which to me are like more clear-cut, distinct semantic fields), then? 

I wish they hadn't used tokens/objects in this example. Wouldn't it be just as clear or clearer if they had stuck to only comparing types/classes?

It may be okay to mix them like that in some contexts, but in other cases it seems like this would be suffering from ignoring/conflating/[better word?] the Type–token distinction.

Does linguistics just not make the https://en.wikipedia.org/wiki/Type%E2%80%93token_distinction ?

This statement seems to reinforce that idea:

words that are examples of categories are hyponyms

because an example of a category/class/type could be either a sub-class or an instance of that category/class/type, right?