Starting with scoring criteria and adding a gimmicky competition isn't even a good idea, never mind the best idea at the conference. Once the criteria are defined, it wouldn't be hard to program a computer to find an optimal solution.

Hirsch notes in the plan: "As Harvard University political scientist Micah Altman has shown, the finite number of potential statewide maps here would be so large as to render the problem intractable not only for current-day computers but probably for any computer ever." Altman, who was also there, has a good PowerPoint presentation that summarizes that research on his website.

Just because the search space is very large doesn't mean you can't get a very good answer in a reasonable amount of time. The number of possible chess games is almost infinte, but computers still play pretty well. My guess is a good mathematician programmer could come up with an answer in a couple of weeks. Possibly it could be improved upon but not by enough to make it worthwhile.