As for screwing up other people's deposits, current information is that it is not very likely. Would you prefer that as a separate exercise I dump more coins into Pirate's scheme directly?
Yes, I would. But you're clearly not going to do that. You'll instead use money from others for it and won't risk your money.
About the net loss: You can throw sand at my eyes all you want, but rest assured you won't blind me
Actually, I am risking my own money - that's what the insurance fund is for. I could easily lose (and expect to at some stage) drop 500 coins in insurance payouts if/when pirate defaults. However, I would prefer that it gets unwound nicely and I don't lose 500 coins in this. I am in it for some profit and it is a good idea making one of the more lucrative schemes available to those who currently do not have access. On that basis, it's a good idea, and one people could easily replicate (low barrier to entry).
Picking up stochastic's points:
Certainly PPT is solvent and can easily meet its obligations - currently we are over-secured (even with tomorrows bond sale)
And also yes the gain is higher than 2% because of the compounding, (but not on the 0.32 that we are putting up - that cannot be invested). As mentioned elsewhere, 1BTC invested for four periods is 1.31 and we're paying 1.28. So 3% plus any premium.
