OC, your phrasing "couldn't fabricate a relatively simple switch." was so deliciously misleading that I had to smile. There was/is absolutely nothing "simple" about the 4 pole thyristor switch impasse. due to the nature of epitaxial structure and the present knowledge of electrical pulsing, the switch they required was, and may still be, beyond the capability of existing science.

The process was shelved because a) POET lacked the huge amount of money required to reach the "undiscovered country" and b) Suresh sensed that no other company was willing to desert silicon sufficiently to make the leap and partner in a new platform

That is why I am fairly confident that their is no secret agreement and/or development with India for the technology either. The semiconductor industry/knowledge/personnel in India is in it's infancy compared to the rest of the semiconductor world. As someone else said "they may have kicked the tires" are realized the tech was not advanced far enough for it to be worth pursuing at this stage of their expertise.

Looking back from here, InP for GaAs was the most judicious pivot they could have made to save POET.

IMO, of course