given the fact that they are expensive and most people cannot do anything else with them in case the bitcoin collapses, and therefore given that they have either a low resale price/are hardly resellable I would say the amount of early adaptors to fpga mining platforms will be limited, until they become significantly cheaper(like below $75). And even then I think that there's only a low amount of people willing to pay for something that they cannot easily resell(like they can do with a GPU)

The fpga miners are a nice idea, but at the current cost the break-even point is over a couple years even with including the electricity savings. The boards do have other interesting potential uses. I hope they do get the price down, but personally don't think I would consider them an option until they get to a price point of 33-25% of what they are now.

There's no way I'm spending $400+ for 100Mh

Or to say this another way:  why should anyone sell effective, let alone super-effective, mining hardware?  It makes more sense for the manufacturer to mine than sell.

None of the currently available FPGA platforms are optimized for cost, primarily since the designers don't know what is required yet. There is also the problem of design by committee which never bodes well.

The only way I see these selling is if they can do 200Mh/s for $100.