This is the reality of novel hardware development and the reason we (usually) don’t announce hardware until it’s ready to ship. In this case, we’ve done a couple of rounds of schematic design reviews with Josh and suggested necessary changes to the design. In the last round of emails, Josh asked a question around PCIe that required us to go upstream and check with suppliers who in turn didn’t respond to us. We then did some further research and provided a response based on our best guess of some possibly underdocumented functionality on the CPU. Each iteration takes time, since for both Josh and for our engineering team, this is one of a large number of other projects that are in flight.
We release documentation and keep interfaces open because we believe in openness, but threads like this explain why many companies don’t.