How about something like this? In the event of a FB IPO within 5 years we look at the valuation of the company 30 days after the IPO and compare it to 33B in todays dollars (with say a 5% IRR).

If facebook goes out of business (unlikely, but want to cover every scenario), you win 10k.

If facebook is purchased outright by another company we take that as a fair valuation and pay the bet out in the same was as an IPO+30 days price.

If facebook is still private 5 years from now it's a push (don't want this to be open ended forever).

We take the difference any multiply it by your "stake" (10k/33B) and the bet pays off.

So, for example let's say facebook IPOs 335 days from today (so IPO+30 = 1 year). With the 5% interest that puts us at 34.7B as the comparison valuation.

So if FB is worth 0 on that day, you win 10k. If FB is worth 20B, you win about 4.5k. If FB is worth exactly 34.7 it's a push. If FB is worth 40B, I win about about 1.6k. If FB is worth 50B, I win about about 4.6k. etc.

1) Make sense? 2) So you in?

Where's Spolsky? He should be in on this. Maybe DHH can put up his Zonda for something that Joel values highly.