I agree it’s a little confusing at first, but I wouldn’t want them to change it. It happens because your original scan file is a negative – the lightest tones of the subject are the darkest tones of the scan and vice-versa. Lightroom inverts the tones when displaying the file so it looks normal, but it’s still a negative – as you’ll find out if you open the scan file in an application that doesn’t respect Lightroom’s adjustments.
One of the guiding principles of Lightroom (and by extension NLP) is that your original file always remains untouched – all the adjustments you can make in Lightroom are non-destructive and can always be undone later if you change your mind or decide on a different interpretation for your image. That’s different from editing in (for example) Photoshop, where once you change the value of a pixel it is changed forever.
The fact that your original file is preserved also means that if Nate someday were to come out with a new version of NLP that did a better job in a certain kind of situation, you could go back and run the new NLP on all your old scan files and get all the benefits of the new version.
The “backwards” behavior of the Lightroom adjustments in an NLP scan make sense as long as you remember that your original scan file is still a negative. I remember that when I was learning to make prints in a traditional “wet” darkroom, I had a hard time getting my head around the concept that if I wanted my print to be darker, I had to give it more exposure on the enlarger, and if I wanted my print to be lighter, I had to give it less exposure. But after a few prints it began to seem more logical to me, and so it will to you!