My hashfast contact and I have been working on the cgminer driver for a while for this protocol, and fortunately since they contacted me early, their final protocol is quite different to what they originally had in mind. I'm very pleased with the design overall, and the cgminer driver for it we've been working on will be pushed to a public repository tomorrow.
The above statement carries a lot of weight in my opinion. I hope recent news dampens some of the FUD in this thread. I'm getting excited by HF now and I'm starting to regret my decision not to place a preorder.
