Models
I have a flawless scientific and philosophical model of reality.
Unfortunately, it's actual size.
I get annoyed when a theorist presents a tractable simplified model and then gets attacked in the questions for leaving bits of the real system out of her description by some audience member who wants to prove they were paying attention. Leaving bits of the real system out of the model is the whole point! If you want to know how the real system behaves, go out into the world and look at it (or run a molecular dynamics simulation on a Beowulf cluster for a few weeks). If you want to understand why it behaves that way, find the simplest, most incomplete model that behaves the same way [1]. Now you know which features of the real system are actually important to its behaviour and which are decorative. That's a pretty good working definition of “understanding”.
[1] Actually, even if it only behaves the same way 90% of the time it's still useful, especially if you know which 90%: it focuses attention on the remaining 10% and on figuring out why they're different.
Unfortunately, it's actual size.
I get annoyed when a theorist presents a tractable simplified model and then gets attacked in the questions for leaving bits of the real system out of her description by some audience member who wants to prove they were paying attention. Leaving bits of the real system out of the model is the whole point! If you want to know how the real system behaves, go out into the world and look at it (or run a molecular dynamics simulation on a Beowulf cluster for a few weeks). If you want to understand why it behaves that way, find the simplest, most incomplete model that behaves the same way [1]. Now you know which features of the real system are actually important to its behaviour and which are decorative. That's a pretty good working definition of “understanding”.
[1] Actually, even if it only behaves the same way 90% of the time it's still useful, especially if you know which 90%: it focuses attention on the remaining 10% and on figuring out why they're different.