I think compared to the very minimal amount that added storage should cost, that this would be a generally unattractive and complex approach. We’ll see what the resource pricing works out as, but I’m expecting the value of safecoin go up while the cost of storage goes down, making it not so attractive to cheat.
The attack vector is the only thing I have attention on along the lines of what you’re saying. I think that that may be a case for smaller initial free cap. That’s a different animal. But even so, I think 250 Mb would be small enough yet give someone enough to be getting on with while starting to use the network.