Will AMD's Compute Abstraction Layer be competitive with Nvidia's CUDA from a usability point of view?

CTM never cut the mustard, it was too close to the metal and thus too difficult for mainstream adoption, but now i hear mention of this mysterious thing called CAL. Anyone got a decent link to some understandable info?

I would like to see AMD do well with their open-source policy, but they seem to be well behind Nvidia on the GP-GPU front:

GP-GPU is starting to matter; I want to be able to accelerate Dirac encoded video using my GPU which is something Nvidia/CUDA can do, but will this ever happen for AMD/CAL?

It's not that it isn't possible for AMD/CAL to do these things, i'm sure it is, but CUDA has got its head start by being easy to use and I have yet to see anything from AMD that suggests their CAL is a competitor to CUDA in this matter..............