This appears to be a continuation of what they have been working on for a lot of years. 

"fusing a sliver of InP onto a silicon wafer, creating a single hybrid semiconductor laser chip"

Problems associated with this approach are the dislocations created in the mismatched matrix across the temperature range.

Yield…since the process does not identify known good die the yield associated with failure of lasers drops as the number of lasers are increased.

Silicon is not the best material for optical routing. Far from it.

Thermal control…Thermal signatures of each device are not isolated as they are in the POET platform where channeling is efficiently applied and POET’s platform is athermal and remains highly accurate across a wide temperature range. 

The design is complex and is the reason it is taking them so long to bring it.