I want to say that the trans world - people who don't identify with their assigned sex/gender - is very different from the gay world - people with a (specific) alternative sexual orientation. This seems obvious (after all they aren't the same thing), but the trouble is, I don't really identify with the trans world, either. I don't see myself as a 'trans person', the way a homosexual might see himself as a 'gay person'.
Maybe it's just that I don't identify with the dominant paradigm of what it means to be trans. Maybe I feel like how a bi person might feel, not quite fitting in with the straights, but not totally fitting in with the gays either. Maybe it's because I feel like I'm transgender, but the trans image focuses on transsexuality. I.e., a trans person is someone who undergoes sexual reassignment surgery (or desires to), and - incidentally - is attracted to the opposite sex they (will or have) become.
In that sense, they are 'straight' transpersons, where a 'gay' transperson would be attracted to the same sex they identify as. Yet, if you're a non-op trans person - which is already unusual, because the trans identity seems to presuppose a desire to change one's body to suit one's mind (maybe we need a 'soft trans', trans with a lowercase 't', to designate gender-crossers who aren't interested in serious sexual solutions like surgery) - those designations flip. If you're attracted to the gender (the mental side of your 'sex') you identify as, you appear straight, because your body's sex is the opposite of the sex of the bodies you are attracted to.
But that presumes you're attracted to a sex and not a gender, or a sex more than a gender. Is it possible that some people are attracted to genders more than sex? What about a bi person who prefers persons who are feminine (or masculine) regardless of their physical sex? How would you define their orientation? People like to make sexual orientation all cut and dry, where everyone has a primary orientation based on attraction to physical sexual characteristics. And then everything else more refined is either one's type or - if it's not socially accepted - one's fetish. So one person's preference for blonde hair is a type, but another person's preference for black cock is a fetish.
I'm not totally against labeling certain types as unusual or abnormal in the strict, literal sense, but we can still acknowledge them as unusual types - representing normal sexual functioning merely directed at unusual targets - and not presuppose they are demonstrations of abnormal sexual functioning in and of itself (so, either get rid of the term 'fetish' or, as I would recommend, consider fetishism a normal part of human sexuality). It severely limits our ability to label and understand the vast diversity of human sexuality if we are constantly affixing moral judgments to every discovery we make.
It's true that some people can be more and less attracted to different targets, such that they can function sexually with a partner who does not represent their ideal type. But does this undermine their statement of their sexual orientation? I would be willing to bet a lot of money that a factually heterosexual male (not one who just claims in denial that he's straight, but who actually is straight and not even a little gay) could still be aroused by a homoerotic stimulus.
Because that's just how our physiology works - there's a mental component, absolutely, but there's also a purely physical component that is not entirely under the control of the mind. Being turned on in that situation doesn't necessarily make one gay. A gay person is more likely to enjoy (on something other than a pure physical, sexual level) that experience and actively seek it out (if he is not in denial about his orientation).
What, then, about people who have a specific type, but are capable of mating outside of that type? Are we to say that their orientation, for example, is to women and not blonde women? But don't non-blonde women possess largely the same characteristics outside of hair color that blondes possess, on the average? Yes, it's true that this person is attracted to women - but a specific type of woman. Saying "attraction to women" is not really enough to pin down his sexual orientation because I'm sure there are women out there he would not have any interest in fucking.
Maybe there are people out there so libidinous that they actually would have sex with just about anyone within a broad category like 'adult females' or what have you. But I think a lot of people are also not like this, and I would hazard a guess that it's the majority. So maybe our concept of sexual orientation itself is too simplistic. What does it hurt to try and update our understanding of sexual orientation with better knowledge of how it functions? Especially in a world where the straight, procreation-only paradigm is inadequate to explain the full diversity of the population?
That procreation requires a man and a woman - any virile man with any fertile woman - need not dictate our preferred sexual practices, when anything other than making babies is the intent. And to disparage anything other than the coupling of a man and a woman as perverse is simply uncivilized. It's time we adopted a more evolved understanding of human sexuality. To do otherwise is to hamper the advance of knowledge, undermining the very purpose of study - but, I fear, is something a lot of moral researchers would be keen to do...
In short, we should replace the concept of 'orientation' which presupposes the answer straight/gay/or bi (which already has the problem of equating sex with gender, and not considering the difficulty of labeling transpersons) with the concept of 'who (or even what) you are attracted to'. We can still use the same term, 'orientation', but the idea should be about attraction, in whatever form it takes, not some small collection of claustrophobic boxes as if to say, you must either be a man who likes women, a woman who likes men, a man who likes men, a woman who likes women, or a man or woman who likes both men and women.
It's better than having just the first two options, to be sure, but it's still far too limiting and doesn't reflect the reality of human diversity. We can still fall back on those options for generalizations - I understand that simplification can be very useful - but they should not be the end of the story, and, above all, we should not expect them to be the end of the story.