Depth first versus breadth first.
Supposition: If there were such a thing as a Benet's for the sciences, the discussion on Yahoo/quizbowl wouldn't be happening. Just a thought, but if you had a single book that gave, not necessarily an in-depth guiide to all the sciences, but which covered a lot of material in ALL the sciences, say to the survey course level, while remaining accessible to the layman. You'd have something that people would be able to look at, absorb, and at least judge whether material for a question was good or bad, easy or hard, obscure or common. We could argue about something's comparative value, because there would be a baseline. Now the problem is that no single work like that exists for science. It doesn't even exist for some subdisciplines of science. (To a certain degree it exists for history of science, but that's only going to get you so far, and a lot of science players don't call that science at all. I do somewhat, only because one can learn a lot about science by how it got from discovery A to discovery B.) Adding to this problem, you have the problem of progressions. Most science and technical curricula march along a series of prerequisites and tracks of classes, to a much heavier degree than outside the sciences. So the really interesting things for writing questions about them won't be accessible to the general public. On the other side, those in the humanities may be better able to take a variety of classes, but that very variety ensures that a lot more is considered fair game. When the science major is confronted with this, and is left with perhaps one elective per semester to cover that scope, you can see where that's equally problematic. In both cases, the best elective either side can take to understand the other side is another packet.

That's why I tend to file the science debate under "intractable issues of quiz bowl." It's the same fight I ran into in the early '90s. I guess I got fluid mechanics into the distribution, but I couldn't get people to get the answers right when they write them. (2100-2300, almost never 4000!)