To an extent, I think all questions of classification - that is either "what defines this type of thing?" or "is this thing a member of that category?" - are ultimately meaningless. Whenever you get too deep into one of these questions, you find out that all the particular questions, exceptions and qualities are fluid and difficult to pin down. Is this person gay? Well, people may be attracted to a broad range of other people, including males and females, with different intensity, although ultimately even their genders are more of a scale than a binary proposition, so what you're really asking is whether this person is attracted to other people of a significantly different outward gender expression than their own, and that's before you get into issues of transgenderism, particular sexual behaviour preferences and changes to all of those things over a lifetime or the other factors that contribute to attraction and have nothing to do with gender at all.
And that's just one question of many types of classification questions that may, ultimately, be too vaguely-defined to even answer. The problem is that we like easy answers, dichotomies and being correct. When there is no "correct" because the question is based on wrong assumptions, we need to let go of those assumptions.
Mokalus of Borg
PS - Of course, this makes the world a much more confusing place.
PPS - As is the case with a deeper understanding of anything, really.