6 ms·
This reminds me of Jonathan's card, which was shortly exploited as an 'experiment' and ultimately frozen by Starbucks https://news.ycombinator.com/item?id=2877
by mzahir 13y ago
This reminds me of Jonathan's card, which was shortly exploited as an 'experiment' and ultimately frozen by Starbucks
https://news.ycombinator.com/item?id=2877779 https://news.ycombinator.com/item?id=2877779
- lotharbot 13y agoJonathan's Card had one key characteristic that made it exploitable (even if Sam hadn't been able to transfer money off of it): there was a reliable way of knowing how much money was on the card, and therefore a reliable way of choosing when to spend (or siphon) money from the card to get high value. I'm not aware of a reliable way of knowing that the car in front of you is "paying it forward". That makes me curious: is it possible to exploit by starting a pay-it-forward chain and then circling back around with a bigger order? Would there be a particular number of cars or time of day or order size that would maximize the likelihood of getting your final order paid for, and would it have a positive expectation when accounting for the initial outlay and the time commitment? [Note that I am not advocating actually trying this; I'm satisfied with it as a thought experiment.]