Sula, you have recently raised the important question of what happened to Poet's PDK.   I recall  the speculation around that event, which was supposed to herald much interest in the industry.   I was impressed with the suggestion that factories need not re-tool, that the Poet solution was sort of plug and play. 

Here is a couple of  quotes from the past.   From:   http://www.poet-technologies.com/news/poet-technologies-and-synopsys-collaborate-on-advanced-modeling-of-planar-electrical-technology-and-development-of-poets-first-process-design-kit.html

The PET PDK and process offers lower cost and simpler process fab options for applications that do not require the full POET optical feature set. Due to the high mobility inherent in III-V materials, PET technology is predicted to deliver performance, which could be equivalent to 3 to 4 nodes ahead of mainstream technologies. Further performance and novel capabilities will be enabled by the incorporation of in-plane optical intra and inter-chip signaling capabilities within the electrical technology. The PET/PDK is scheduled to be available at the end of Q4 2014 and will allow POET to provide detailed design information to industry fab partners and customers. This will enable pre-semiconductor design evaluation to integrate optical, analog and digital functions together. PET-based electronic devices represent a breakthrough in performance and power efficiency.

Mr. Copetti concluded, “We are delighted to target a 40-nm technology node for our first PDK delivery. This is an aggressive target for our POET devices and process IP. This will demonstrate to the industry the significance and the strengths of POET compared to the performance and power efficiency of today’s widely used technologies.”

Here is another:  From:   https://seekingalpha.com/article/3278165-poet-technologies-about-to-move-from-stealth-mode-to-wealth-mode

The potential is for POET to be both complementary and disruptive to the $600 billion semiconductor industry. It's an elegant solution to the growing complexities associated with improvement within standalone silicon-based semiconductors. POET's processes can be readily introduced to current silicon CMOS fabs, saving companies tens of millions of dollars in retooling, upgrades and R&D.

>>If anyone is going to the AGM, this matter about the PDK might be a good question to bring forward.

>>Those were the good old days, .... eh?