Well, the consensus is that the value of an idea is almost certainly zero, so the only way to make it worthwhile is to get one runaway success in a big pool of

Well, the consensus is that the value of an idea is almost certainly zero, so the only way to make it worthwhile is to get one runaway success in a big pool of attempts. Everyone knows the VC version of this approach. This is how it would look like for a programmer. I'm not dead-set on these specific upfront costs and equity split, this is just the reverse of the original offer.