Fiber has fairly narrow windows in which it is as transparent as it needs to be to go long distance. We're already pretty good at filling these windows with conventional semiconductor lasers.
What this is actually interesting for is being able to access arbitrary atomic transitions, many of which are outside the range of conventional semiconductors (too short, usually - there's a big hole between green and red for semiconductors). That's why they talk about quantum stuff.
This is true. But even within this window, e.g. between 1100 nm infrared and 700 nm red, we could put 40 different "colors" at 10 nm steps. Separation at the receiving end may become hard though.
Standard ITU grid is 100 GHz channel spacing, with subdivisions of 25 and 50. We're routinely using symbol rates high enough that the channels are fairly well filled.
What this is actually interesting for is being able to access arbitrary atomic transitions, many of which are outside the range of conventional semiconductors (too short, usually - there's a big hole between green and red for semiconductors). That's why they talk about quantum stuff.