Title text: To be fair, the braised and confused newt on a bed of crushed Doritos turned out to be delicious.
A genetic algorithm starts with a set of candidates and evaluates them. The best candidates are combined and randomly mutated to form the candidates for the next generation. After being allowed to proceed for an extended period, a genetic algorithm can often produce remarkable results. If the initial candidates are randomly-generated (as appears to be the case here), the initial generations are usually horrible.
In the comic, the computer science department is the host of a dinner party based on recipes produces by a genetic algorithm. Based on the remarks of the second diner, this is probably not the first generation, and the results are still horrible. The host of the party is so enamored of the promise of the genetic algorithm that he fails to take into account that it will be several years before the recipes become remotely good.
- [Three people, one woman and two men, sit along a table with dishes and drinks in front of them. A fourth man is walking in, a plate with food on it in one hand, a laptop in the other.]
- Woman: I’ve got... Cheerios with a shot of vermouth.
- Man #1: At least it’s better than the quail eggs in whipped cream and MSG from last time.
- Man #2: Are these Skittles deep-fried?
- Cueball: C’mon guys, be patient. In a few hundred more meals, the genetic algorithm should catch up to existing recipes and start to optimize.
- We’ve decided to drop the CS department from our weekly dinner party hosting rotation.