So the basis of your argument is that ASICMiner is a miracle, and all the other companies are going to go the way of BFL/Avalon?
That's a bold gamble, considering the starting cash that some of these guys are getting (Cointerra, for example, has experienced industry professionals and $1.5M). And the "2nd or 3rd gen tech" argument is silly, since the new startups are going straight for those nodes, they're not starting at gen 1.
The reality is that the barrier to entry for new ASIC companies is quite low. In my opinion, either ASICMiner was a lighting strike, or things are going to turn sour for them.
But hey, I've been saying this since 2.5 BTC a share. I'm wrong so far.
ASICMiner is ahead to this day and thus in superior position.
But as You say, boo can happen if they fall asleep or slow down.
Currently I don't see any problem since everything is going according to the plan.
Even rise of the competition is anticipated already.
My point is that the concept of a company being "ahead" in market positioning is being misapplied to this situation. Customers have little brand loyalty in the bitcoin miner space - it's all about the best cost to GH/s and power ratio. And the ability to deliver, of course - but that is easy to demonstrate if you've got the hardware.
ASICMiner is developing next-gen 28nm chips - but so are the competition, including people who have much more experience in ASIC design than AM (Cointerra is my bet here). That will effectively erase the lead that AM has built, which is not something that typically happens with established companies. AM has not been around long enough to have any "lead" to speak of.
In Cointerra's case, the fact that they have private equity means that they probably have a real, honest-to-god plan for their supply chain. They haven't already raked in the cash from pre-orders, so the success of their business is completely dependent on shipping units. Their investors knew this and still dumped a very generous amount of cash into their company. Odds are that their angel investors are going to make absolutely sure that they have the resources necessary to succeed.
28nm tech already exists, and is open to these companies for use. AM has undoubtedly gained lessons in design from their last iteration, but Cointerra brings an overwhelming amount of practical experience out the door. Check these bios - seriously.
http://www.cointerra.com/team/The ASIC market is waking up. My advice: do not assume it will continue to belong to AM. That is a risky bet you will barely profit from winning, and you will greatly suffer from losing.