I don't have a problem with labels. I have a problem with labels that are murky and confusing.

Let's go for a little more clarity here. We're not talking about labels, we're talking about concepts. A label can be as particular as you like, naming an exact individual: Sophie, Sarah, Reine, etc. Concepts are general. They are meant to gather individual entities into a mental grouping in order to make them easier to think about. They are essential for human thought. We could not possibly deal with every single apple as if it were an entirely new phenomenon completely unrelated to all the other apples we've ever seen in our life. Multiply that by all the gazillion entities out there in the universe, and you can easily see how our brains would get nowhere trying to handle all that date if we did not form concepts.

So yes, concepts are good, they're essential. They only work properly, though, when they're formed properly. There are two important steps in concept formation. One is to make sure that the items being grouped have an essential similarity that makes them cognitively functional. One can't form a concept that groups tires with pencils. There's nothing we can do, cognitively, with that concept. We can group apples, cherries, lemons, mangos, etc, under the concept 'fruit', but tires & pencils is a non-starter. I'll gloss over the question of whether transsexuals and crossdressers are cognitively commensurate, and just stipulate that on the basis of gender non-conformity, they can be grouped together.

The other important step when forming a concept is to give it a name. If the name is made up from whole-cloth, then no problem. Where in the world did the word 'fruit' come from? Who knows? But someone came up with it, and it stuck. Scientists like to invent words out of Latin and Greek roots. When they do so, those roots should be employed properly. If they describe a creature as 'monocephalic', it should have just one head. People who know what the root words mean are going to expect that, and it would be confusing to call something with two heads monocephalic.

The naming schemes they come up with should also be internally consistent. For example, if scientists doing research in human sexuality want to come up with a name for heterosexuals who have occasional sexual fantasies about members of the same sex, it would be a grave error to call them 'polysexuals'. Sure, they can stipulate their own definition of the word, and give it the weight of expert authority. Given current usage of the words 'homosexual', 'heterosexual' and 'bisexual', however, people will misunderstand that word. They will take it to be a statement about the person's sexual preference, not about his sexual fantasies. The people saddled with that "label" will constantly have to explain that it doesn't mean what it appears to mean; or they will simply reject it--perhaps by saying they hate labels.

In my opinion, there's nothing wrong with the underlying concept that groups people together by the fact that they rebel against gender conformity. The problem I have is with the name that has been given to the concept. The similarity in form of 'transgender' and 'transsexual' leads people to think that they are synonyms. There is simply no reason to read the 'trans' in 'transgender' differently from the 'trans' in 'transsexual'. The two words have the exact same construction, and should be read the exact same way. When transsexuals decide that they can be happy with changing their sex just for a few hours every other week, then I'll believe that the 'trans' in 'transgender' means "crossing back and forth," rather than crossing over permanently.