quoting appendix N.1 'New Features' from the OpenGL 3.0 spec:
"Half-float (16-bit) vertex array and pixel data (GL_NV_half_float and GL_ARB_half_float_pixel)."
edit: added the underscores
This probably includes adding a lot of functions. It might be useful to have the two extension specifications at hand
FWIW- I have an implementation of Half as an actual haskell numeric data type that sits in my openexr bindings:
It currently cheats and uses the OpenEXR conversions to/from Float, but those could be written by hand.
If we're going to bind out to Half's it'd be nice to have a data type that let you duplicate their behavior CPU side.